Feb 25, 2005 - load and that may lead to erosion and thus a loss of carbon, see below. 3. Hydrological ...... human destruction of forest since the last glaciation led to the formation of blanket ...... mean July water table, and site coded as dummy variables. Axes 1 ...... Global dimming (Wild et al., 2005) may also have some.
The Influence of Management on the Vegetation and Carbon Fluxes of Blanket Bog
A thesis submitted for the degree of Doctor of Philosophy University of Edinburgh Alan Gray March 2006
i
Declaration This thesis was written by myself and represents the results of my own work, except where stated otherwise, and has not been submitted in an application for any other degree.
Alan Gray, March 2006
ii
Acknowledgements This research was funded by the Natural Environment Research Council from 2001 until 2004 (Award number NER/S/A/2001/06164) with additional support from the Royal Society for the Protection of Birds and in 2005 by The Scottish Executive, Scottish Natural Heritage and the Royal Society for the Protection of Birds. I would particularly like to thank Norrie Russell, James Plowman Alisdair Boulden and Katy Robinson of the RSPB at Forsinard for providing accommodation and much needed conversation during field work as well as invaluable fire extinguishing services. The RSPB’s Clifton Bain and Neil Cowie deserve special mention not only for making comments on drafts but also as Clifton not only sought but secured extra funding for 2005. Information for Chapter 1 came from various sources but I would like to thank Andrew Coupar for answering my many and varied question on UK peatlands, Paul Robinson for information on the Countryside Survey 2000, Kate Heal for sources of information on DOC etc. in river environments, and I am particularly grateful for written contributions by Neil Wilkie concerning the Peatland Management Scheme, LIFE Nature and Heritage Lottery sections, Mike Wood for the Scottish Forestry Grants Scheme, and Mandy Gloyer for Agri-environment Schemes and Land Management Contracts. The schematic drawing of the peat carbon cycle in Chapters 1 and 6 was inspired by a leaflet by Dartmoor National Park Authority available at: http://www.dartmoornpa.gov.uk. I am grateful to the British Atmospheric Data Centre, which provided access to the Met Office Land Surface Observation Station Data at Kinbrace. Nick Ketteridge and Andy Baird from Sheffield University for access to Nick’s climate data from Maol Donn during July to September 2004 and Brian Hart for providing temperature and rainfall data from Dalhalvaig for 2004. John Adamson at the Environmental Change Network was very helpful with access to Moor House and data from the Hard Hill experiment, which I was unfortunately iii
unable to put to use for this thesis because of hard disk failure and time constraints having to resort to my own smaller dataset. Thanks to Bruce Ball at the Scottish Agricultural College for the loan and instruction of a soil penetrometer. Thanks to David Sales who stepped in at the last minute and gave statistical advice on how to analyse my unbalanced and peculiar experimental design. Rui Zhang was very helpful with advice on the non-rectangular hyperbola and implementation of the equations using solver in Excel. Dr. Colin Legg, whose door was never closed, was particularly supportive throughout the entire PhD from the initial conception of the thesis through to correcting various chapter and report drafts, without his help this thesis would be much the poorer. Professor John Grace also made invaluable comments on draft chapters and together with Professor Keith Smith gave much needed advice particularly concerning the measuring of carbon dioxide and methane fluxes and on experimental and technical design. I would also like to thank Franz Conen, Rab Douglas, Dave Reay and Karen Dobbie for help and advice on using the gas chromatograph and field equipment. John Blane needs a mention for teaching me that there was more than one grass species and encouraging me to go to University. I would particularly like to thank my friends and family particularly my Mum and Dad for being supportive throughout the entire period, always ready with a consoling chilled beer and for sitting us down in front of the likes of David Attenborough, and Jacques Cousteau, providing inspiration for investigation of the natural world. Sam Gardner and Paul Robinson also deserve mention for their ability to listen to someone moan about their PhD for the umpteenth time, even when doing PhD’s themselves. Finally I would like to thank Claire Pannell for her enduring ability to put up with me through what at some points can only be described as hell and still be there with a smile, thank you. iv
Abstract This thesis presents evidence of the impact of anthropogenic management on the blanket bog ecosystem. The effects of management on carbon fluxes and vegetation through control of grazing and burning for blanket peats in the UK are explored and calculations of tentative climate warming potential of sample sites in the Sutherland and Caithness peatlands are presented. An examination through semi-quantitative literature review and the analysis of published field work data, of the relationship between the management of blanket bog and gaseous carbon fluxes in the UK, is presented. The geographical distribution of peatlands and blanket bog in the UK and the management actions that influence them are summarized. Previous work in relation to management on blanket bog is reviewed and some hypothetical ways in which management may affect carbon fluxes are discussed. The main published works in the UK on carbon flux from peatland systems is reviewed, including fluxes to river systems in the form of dissolved organic carbon. A semi-quantitative synthesis of the published gaseous carbon fluxes in the UK reveals gaps in research. Mean methane emission is approximately 0.029 µmol CH4 m-2 s-1, but there is no reliable estimate for net gaseous flux rates of carbon dioxide from UK blanket peats and both winter fluxes and the impact of peatland management practices have been understudied. The links between vegetation and management are analyzed through vegetation survey of blanket bog areas in the Caithness and Sutherland peatlands at the RSPB Forsinard Reserve and a long-term burning and grazing split plot experiment in the Moor House-Upper Teesdale National Nature Reserve. Vegetation structure as well as species composition was shown to be affected by management. The National Vegetation Classification method was insensitive to management treatments and may be of limited use for indicating management practices in the wider landscape for peatland ecosystems in the UK. Key climatic controls of gaseous carbon fluxes at the site scale were photosynthetically active radiation for CO2 in the light, soil temperature for CO2 in the dark and soil temperature for CH4 flux. There were some departures from theoretical predictors of gaseous fluxes that may have links to site management. The influence of management on the gaseous fluxes from the blanket bogs of the RSPB reserve at Forsinard is explored through the use of general linear models and v
regression analyses. A tentative carbon balance for certain sites within the reserve over the period of a year indicates that differences between sites that may be attributed to management. Heavily damaged sites appear to fix less and respire more CO2. Fire may lead to initial increase in CH4 emissions. However, the effect of management in terms of drainage may not always be immediately apparent. Further temporal and spatial resolution of the effects of peatland management on carbon fluxes is required. Proposal for further research include the calibration of indicators of carbon fluxes in UK peatlands.
vi
Preface The majority of PhD research has its problems. I, like many other students before, have had my fair share of ‘ordinary’ problems such as, leaking flux chambers leading to months of redesign and testing, faulty datalogger’s that take 3 months to fix, experiments that take weeks to set up then don’t work, and even weather that appears to conspire to disrupt field work. These though dreadful at the time fade once a sufficient period has passed, or writing up commences, only to be recalled during discussions with friends over several pints of beer. However I had two rather more serious problems, which help to put this thesis in context. The first of these happened on the 13th (I’m not making this up honest!) of May 2004 and is illustrated below. No I did not do my field work in a war zone, but yes this was my Land Rover. The fire, caused by a faulty starter motor, also set the adjacent Forest on fire and without the timely intervention of James Plowman would have been far more serious. This resulted in a brief period where communication with other people was difficult due to my propensity for blasphemy during normal conversation, but I did recover. More importantly the loss of equipment put field work back 2 months while the equipment and the vehicle were replaced (which is another long story involving Land Rovers). The second problem happened at the tail end of 2004 and is more commonplace, hard disk failure. This is bad enough but was compounded by all of my back up disks being corrupted and resulted in the loss of 6 months work. I was saved from complete disaster by the timely intervention of the RSPB who together with the Scottish Executive and SNH funded my research for an extra year in 2005. So remember; There is no such thing as having too many back ups, and be wary when buying Land Rovers.
vii
There so much I wanted to say here but in the end the photograph says it all!!
viii
Thesis Aim and Layout Thesis Aim The general aim of this thesis is to examine the influence of management practices on the gaseous carbon fluxes of blanket bog. Thesis Layout The layout of this thesis is generally in the style of scientific papers, although this differs in Chapters 1, 6, and 7. However in an attempt to avoid unnecessary repetition, some introductions are shorter than normal and where methods have already been detailed subsequent chapters will only refer to the previous chapter where the methods are already stated. Versions of Chapters 1, 2 and 5 appeared as unpublished reports for the Scottish Executive and the RSPB in 2005. Chapter 1: Introduction. This summarizes the geographical distribution of peatlands and blanket bog in the UK and the management actions that are carried out on them. Previous work in relation to management on blanket bog is reviewed and some hypothetical ways in which management may affect carbon fluxes is discussed. A review of some of the main work in the UK on carbon flux from peatland systems and from peatland river systems in the form of dissolved organic carbon is also included. Chapter 2: Peatland gaseous carbon fluxes and land management: searching for a paradigm. The main work on gaseous carbon fluxes in the UK is semi-quantitatively reviewed and an attempt to synthesize previous work to identify areas of future research is made. Chapter 3: Blanket Bog Site Characteristics and the Role of Management The vegetation of blanket bog areas belonging to the Caithness and Sutherland Peatlands within the RSPB Reserve at Forsinard are described and analysed in relation to management and site specific factors. A vegetation survey of a split plot burning and grazing experiment is also analysed to determine how this type of management affects blanket bog vegetation.
ix
Chapter 4: Environmental relationships to the gaseous carbon fluxes of blanket bog. Gaseous flux data from the blanket bogs of the RSPB reserve at Forsinard are used to identify the main environmental climatic controls through the use of regression models. Chapter 5: Does management influence the gaseous carbon fluxes of blanket bog? The influence of management on the gaseous fluxes from the blanket bogs of the RSPB reserve at Forsinard is explored through the use of General Linear Models. Regression models are also used to explore a tentative carbon balance for certain sites within the reserve over the period of a year. Chapter 6: Discussion. This discussion brings the previous chapters together and discusses what the thesis means as a whole. Chapter 7: Conclusions and Further Work. Summary concluding points are made from all chapters and suggestions for further research are made.
x
Table of Contents
Page
Declaration
i
Acknowledgements
ii
Abstract
iv
Preface
vi
Thesis Aim and Layout
viii
Chapter 1
Blanket Bog Ecosystem Carbon Fluxes and Management
1
1.1
Introduction
1
1.2
Geographical Extent of Blanket Bog and Management Relevant to Scotland
6
1.2.1
Total Blanket Bog Resource
6
1.2.2
Geographical Extent of Grazing
11
1.2.3
Geographical Extent of Burning
11
1.2.4
Geographical Extent of Drainage
11
1.2.5
Geographical Extent of Erosion
11
1.2.6
Geographical Extent of Peat Extraction
12
1.2.7
Geographical Extent of Conservation
12
1.2.8
Geographical Extent of Restoration
12
1.3
Carbon Cycle of Blanket Bog
13
1.3.1
Carbon Dioxide
15
1.3.2
Methane
16
1.3.3
Peatland Carbon Fluxes to River Systems
17
1.4
UK Peatlands and the Greenhouse Gas Inventory (GGI)
22
1.4.1
GGI and Unaccounted Emissions from Peatland
25
xi
Contents
Page
1.5.1
Grazing
26
1.5.2
Hypothesized Effect of Grazing on the Carbon Balance of Blanket Bog
29
1.5.3
Burning
30
1.5.4
Hypothesized Effect of Burning on the Carbon Balance of Blanket Bog
35
1.5.5
Drainage
36
1.5.6
Hypothesized Effect of Drainage on the Carbon Balance of Blanket Bog
39
1.5.7
Erosion
39
1.5.8
Hypothesized Effect of Erosion on the Carbon Balance of Blanket Bog
42
1.5.9
Peat Extraction
42
1.5.10
Effect of Peat Extraction on the Carbon Balance of Blanket Bog
42
1.5.11
Conservation
42
1.5.12
Hypothesized Effect of Conservation on the Carbon Balance of Blanket Bog
43
1.5.13
Restoration
43
1.5.14
Hypothesized Effect of Restoration on the Carbon Balance of Blanket Bog
44
1.6
Examination of Policy Mechanisms for Blanket Bog Restoration
45
1.7
Conclusions
48
1.8
References
48
xii
Contents
Page
Chapter 2
Peatland gaseous carbon fluxes and land management: searching for a paradigm.
63
2.1
Introduction
63
2.2
Methods
64
2.3
Results
66
2.4
Discussion
72
2.4.1
Fluxes of CO2
72
2.4.2
Fluxes of CH4
73
2.4.3
UK Peatlands, overall C source or sink?
73
2.4.4
Representation of sampled sites
74
2.4.5
Management
75
2.4.6
Climate change: models, ecosystem response and government policy
76
2.4.7
Peatland carbon flux research: opportunities
77
2.5
Conclusions
78
2.6
References
78
Chapter 3
Blanket Bog Site Characteristics and the Role of Management
86
3.1
Introduction
86
3.2
Study aims
86
3.3
Methods
87
3.3.1
Site Descriptions
87
3.3.1a
Moor House
87
3.3.1b
Forsinard
88
3.3.2
Vegetation characterisation
91
xiii
Contents
Page
3.3.2a
Moor House
91
3.3.2b
Forsinard
91
3.3.3
Statistical Analyses
92
3.3.4
Community comparison
93
3.4
Results
96
3.4.1
Moor House
96
3.4.2
Forsinard
104
3.5
Discussion
121
3.5.1
Management effects on vegetation experimental evidence
121
3.5.2
Management effects on vegetation evidence from Forsinard
123
3.6
Conclusions
125
3.7
References
126
Chapter 4
Environmental relationships to the gaseous carbon fluxes of blanket bog
131
4.1
Introduction
131
4.2
Study aims
132
4.3
Methods
132
4.3.1
Site Description
132
4.3.2
Gas Flux Measurements
132
4.3.3
Statistical Analyses
135
4.4
Results
137
4.5
Discussion
154
4.6
Conclusions
158
4.7
References
158
xiv
Contents
Page
Chapter 5
Does management influence the gaseous carbon fluxes of blanket bog?
165
5.1
Introduction
165
5.2
Study aims
166
5.3
Methods
166
5.3.1
Site Description
166
5.3.2
Vegetation Characterisation
166
5.3.3
Gas Flux Measurements
166
5.3.4
Statistical Analyses
167
5.4
Results
169
5.4.1
Management effects on gaseous carbon fluxes
169
5.4.1a
Statistical analyses using GLM
169
5.4.1b
Graphical exploration of Main Sites and Sites L, M and N
171
5.4.2
Relationship of vegetation to water table penetrometer data and gas flux responses slopes
175
5.4.3
Tentative carbon balances
177
5.5
Discussion
184
5.5.1a
Statistical analyses using GLM
184
5.5.1b
Graphical exploration of Main Sites and Sites L, M and N
186
5.5.2
Relationship of vegetation to water table penetrometer data and gas flux responses slopes
187
5.5.3
Tentative carbon balances
187
5.5.4
UK Climate Change Scenarios
191
5.6
Conclusions
193
5.7
References
194
xv
Contents
Page
Chapter 6
Discussion
201
6.1
Critique of Flux Chamber Methodology
201
6.1.1
Chamber Critique: Theoretical Considerations
201
6.1.2
Chamber Critique: Chamber Design and Alteration of Ambient Conditions
203
6.1.3
Chamber Critique: Disturbance associated with base insertion
206
6.1.4
Chamber Critique: Other Methodological Noise
208
6.2
Does Management Affect Carbon Fluxes?
209
6.3 Chapter 7
References Conclusions and Further Research
213 217
7.1
Conclusions
217
7.2
Further Research
220
7.3
References
222 224
Appendices 8.1
Chapter 1 Appendix
224
8.2
Chapter 2 Appendix
233
8.3
Chapter 3 Appendix
242
8.4
Chapter 5 Appendix
243
8.4.1
Minitab GLM Output
244
8.4.1a
Main Sites 2003-4 CO2 Light Flux
245
8.4.1b
Main sites 2003-4 CO2 Dark Flux
253
8.4.1c
Main sites 2003-4 CH4 Flux
256
8.4.1d
Main Sites 2005 CO2 Light Flux
260
xvi
Contents
Page
8.4.1e
Main sites 2005 CO2 Dark Flux
261
8.4.1f
Main sites 2005 CH4 Flux
270
8.4.1g
Fire sites CO2 Light Flux
275
8.4.1h
Fire sites CO2 Dark Flux
280
8.4.1i
Fire sites CH4 Flux
282
8.4.1j
Drain sites CO2 Light Flux
286
8.4.1k
Drain sites CO2 Dark Flux
291
8.4.1l
Drain sites CH4 Flux
293
8.5
Appendices References
296
xvii
Table of Figures
Page
Chapter 1 Figure 1.1:
The rising concentrations of CO2 recorded at Mauna Loa
2
Figure 1.2:
The soil carbon content of the United Kingdom
4
Figure 1.3:
Schematic representation of the peatland carbon cycle
Figure 1.4:
Simplified successional changes between bog and heath communities
28
Figure 1.5:
Artificial moor-gripping network on blanket peat near Forsinard, Sutherland in Scotland.
37
Figure 1.6:
A simplified scheme of bog degradation and erosion
41
Figure 2.1:
Carbon dioxide flux against study number
69
Figure 2.2:
Site mean methane flux results from published papers
71
Figure 2.3:
Model relating CH4/CO2 emission ratio to Global Warming Potential and time for UK peatlands
72
Figure 3.1:
Locations of Hard Hill experimental plots and location of Moor House
87
Figure 3.2:
Details of Hard Hill experimental set up
88
Figure 3.3:
Locations of sampling sites in relation to Forsinard Sutherland
89
Figure 3.4:
Axes 1 and 2 of a DCA of species percentage cover data showing samples from Hard Hill
97
Figure 3.5:
Axes 1 and 2 of a DCA of species percentage cover data showing species from Hard Hill
98
Figure 3.6:
Axes 1 and 2 of an RDA of species percentage cover data against site treatment from Hard Hill
99
14
Chapter 2
Chapter 3
xviii
Figure
Page
Figure 3.7:
Axes 1 and 2 of PCA plots of Hard Hill samples and Eddy et al (1969) Calluneto-Eriophoretum communities and contribution of NVC communities to the PCA ordination subspace
103
Figure 3.9:
Boxplots of pH by site from the Forsinard reserve
104
Figure 3.10:
Mean (+/- SE) penetrometer readings (k Pa) every cm for 7 sites in the Forsinard reserve
105
Figure 3.11:
Mean (+/- SE) penetrometer readings (k Pa) per cm depth for 10, distance from drain, transects from the Cross Lochs Drain site
106
Figure 3.12:
Mean (+/- SE) penetrometer readings (k Pa) every cm depth for the unburnt and burnt sites in the Forsinard reserve
107
Figure 3.13:
Axes 1 and 2 of DCA of samples from Forsinard vegetation relevés
113
Figure 3.14:
Axes 1 and 2 of DCA of species from Forsinard vegetation relevés
114
Figure 3.15:
Axes 1 and 2 of CCA sample plot, of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve
116
Figure 3.16:
Axes 1 and 2 CCA species plot, of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve
117
Figure 3.17:
Axes 1 and 2 of CCA sample plot, of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve
118
Figure 3.18:
Axes 1 and 2 of CCA species plot, of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve
119
Figure 3.19:
Axes 1 and 2 of RDA of PObscured data for vegetation plots with sites as nominal explanatory variables
120
Figure 4.1:
Light (left) and dark (right) chambers used for the measurement of CO2 and CH4 fluxes from blanket bog at Forsinard 2003-2005
135
Figure 4.2:
CO2 light non rectangular hyperbola, light response curve by site.
140
Chapter 4
xix
Figure
Page
Figure 4.3:
CO2 light log linear regression by site.
143
Figure 4.4:
CO2 dark response linear regression by site.
146
Figure 4.5:
CO2 dark exponential regression by site
149
Figure 4.6:
CH4 soil temperature linear regression by site
151
Figure 4.7:
CH4 soil temperature exponential regression by site
153
Figure 5.1:
Mean (+/- SE); (a) carbon dioxide flux in the light (µmol CO2 m-2 s-1) (b) carbon dioxide flux in the dark (µmol CO2 m-2 s-1) and (c) methane flux (µmol CH4 m-2 s-1), at main sites and site L in July 2004.
171
Figure 5.2:
Mean (+/- SE); (a) water table (mm), (b) soil temperature (oC) at main sites and site L in July 2004, and (c) global solar irradiation (kJ m-2) and (d) air temperature (oC) from Kinbrace Weather Station
172
Figure 5.3:
Mean (+/- SE); (a) carbon dioxide flux in the light (µmol CO2 m-2 s-1) (b) carbon dioxide flux in the dark (µmol CO2 m-2 s-1) and (c) methane flux (µmol CH4 m-2 s-1), at main sites and site L in July 2004
173
Figure 5.4:
Mean (+/- SE); (a) water table (mm), (b) soil temperature (oC) at main sites and site L in August 2004, and (c) global solar irradiation (kJ m-2) and (d) air temperature (oC) from Kinbrace Weather Station
174
Figure 5.5:
Axes 1 and 2 of CCA of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve showing samples
176
Figure 5.6:
Axes 1 and 2 of CCA of species percentage cover data from plots used for gaseous flux measurements in nine peatland sites on the RSPB Forsinard Reserve showing species
177
Figure 5.7:
CO2 light non-rectangular hyperbola responses by site
178
Figure 5.8:
CO2 light log linear regression responses by site
178
Figure 5.9:
CO2 in the dark soil temperature regression responses by site
179
Chapter 5
xx
Figure
Page
Figure 5.10:
Methane soil temperature regression responses curve by site
179
Figure 5.11:
Modelled sum of carbon dioxide fluxes for Forsinard sites using linear models
180
Figure 5.12:
Modelled sum of carbon dioxide fluxes for Forsinard sites using non-rectangular hyperbola
181
Figure 5.13:
Modelled sum of carbon dioxide and methane fluxes for Forsinard sites using linear models
183
Figure 5.14:
Modelled sum of carbon dioxide and methane fluxes for Forsinard sites using nonrectangular hyperbola model
184
Figure 6.1:
The effect of inserting chamber base into Calluna dominated bog.
207
Figure 6.2:
Schematic representation of the peatland carbon cycle with the influence of management superimposed
210
CH3 App. Figure 1:
Axes 1 and 2 of DCA of samples from Forsinard vegetation relevés with (a) all samples and (b) only gas flux samples
242
CH3 App. Figure 2:
Deer and sheep footprints mapped across the Forsinard and Dorrery reserve.
243
CH5 Appendix Figure 1:
Forsinard daily mean PAR versus modelled daily PAR for the same days in 2004.
244
CH5 App Figure 2: CH5 App Figure 3: CH5 App Figure 4: CH5 App Figure 5: CH5 App Figure 6: CH5 App Figure 7:
Residual plots for Main sites 2003-4 CO2 Light Flux
245
Residual plots for Main sites 2003-4 CO2 Dark Flux
253
Residual plots for Main sites 2003-4 CH4 Flux
256
Residual plots for Main sites 2005 CO2 Light Flux
260
Residual plots for Main sites 2005 CO2 Dark Flux
267
Residual plots for Main sites 2005 CH4 Flux
270
Chapter 6
Appendices
xxi
Figure CH5 App Figure 8: CH5 App Figure 9: CH5 App Figure 10: CH5 App Figure 11: CH5 App Figure 12: CH5 App Figure 13:
Page Residual plots for Fire sites CO2 Light Flux
275
Residual plots for Fire sites CO2 Dark Flux
280
Residual plots for Fire sites CH4 Flux
282
Residual plots for Drain 2003-4 CO2 Light Flux
286
Residual plots for Drain CO2 Dark Flux
291
Residual plots for Drain CH4 Flux
293
xxii
List of Tables
Page
Chapter 1 Table 1.1:
Estimates of the peatland resource of Scotland
9
Table 1.2:
Estimates for the geographical extent of management on blanket bog in Scotland
10
Table 1.3:
Summary of sources and sinks of greenhouse gases important to UK peatlands for the year 2002
23
Table 1.4:
Average soil carbon density for different land cover in the UK
24
Table 1.5:
Activity and Emission Factor Data for Upland Drainage
24
Table 1.6:
Breakdown of the contribution of upland drainage and peat extraction to subcategory 'other'
24
Table 1.7:
Summary of impacts of burning management on selected blanket bog species, species groups and blanket bog habitat
32
Table 1.8:
Extent of the practice of burning and advantages and disadvantages of this type of management on the blanket bog habitat
34
Table 2.1:
Examples of keywords used in literature review searches
65
Table 2.2:
Number and characteristics of gaseous CO2 and CH4 flux studies conducted in the UK from a review of papers
66
Table 2.3:
Mean carbon dioxide flux results from published papers
68
Table 2.4:
Site mean methane flux results from published papers
70
Table 3.1:
Details of Hard Hill Site at Moor House NNR and the 11 sampling sites located within the Forsinard Reserve
90
Table 3.2:
Number of relevés per site and dates of vegetation sampling from Forsinard sites 2004-2005
91
Chapter 2
Chapter 3
xxiii
Table
Page 96
Table 3.3:
Species, Species code, and total number of relevés in each treatment for each species recorded from a total of 72 relevés sampled from Hard Hill experimental site, Moor House NNR
Table 3.4:
Community comparison of relevés data in particular experimental treatments with NVC and Calluneto-Eriophoretum
100
Table 3.5:
Percentage species match expressed as a percentage of the species found in community row with community column and number of specie in each community
102
Table 3.6:
The specie faecal count and number of footprints found in 72 relevés for each of the sites sampled at Forsinard reserve.
109
Table 3.7:
Species, Species code, site presence and total number of relevés for each species recorded from total of 185 relevés from Forsinard and Dorrery Nature Reserve
110
Table 4.1:
Number of plots, plot codes and dates of gas flux and vegetation sampling from Forsinard sites 2003-2005
134
Table 4.2:
Results of stepwise regression of net CO2 light flux and CH4 flux and climate variables from selected sites located in the Forsinard Reserve
137
Table 4.3:
Regression equations for carbon dioxide fluxes in the light and dark and methane flux with associate R2 adjusted, p values and degrees of freedom
138
Table 5.1:
General linear model terms
168
Table 5.2:
Summary of effects and interactions analysed using general linear models for CO2 light, CO2 dark and CH4 fluxes with associated p values and degrees of freedom (df) for sites in the Forsinard and Dorrery Nature Reserve
169
Table 5.3:
Maximum, minimum and mean monthly temperature, hours of sunshine and rainfall for the North of Scotland for 2004
189
Chapter 4
Chapter 5
xxiv
Table
Page Predicted seasonal changes in temperature and rainfall by 2080 using UKCIP02 high and low emission scenarios with respect to model-simulated 1961-1990 climate
193
CH1 App. Table 1:
Relationship between mole and mass in grams of chemical substances relevant to this thesis
224
CH1 App. Table 2:
Prefixes and multiplication factors in common use.
224
CH1 App. Table 3:
Carbon fluxes and concentrations in rivers in the UK from peatland catchments.
227
CH1 App. Table 4:
Management and soil characteristics for studies in CH1 Appendix Table 3
232
CH2 App. Table 1: CH2 App. Table 2:
Published fluxes of methane from research on peatlands in the UK.
234
Mean methane flux results from published papers examined by this thesis
237
CH2 App. Table 3:
Carbon dioxide fluxes in common units from reviewed sources.
238
CH2 App. Table 4:
Methane fluxes in common units from reviewed sources.
239
CH5 App. Table 1:
Missing value ordinary least squares regression model equations
243
Table 5.4:
Appendices
xxv
Chapter 1 Chapter 1: Blanket Bog Ecosystem Carbon Fluxes and Management This chapter introduces the blanket bog as an ecosystem and places it within a UK and Scottish perspective as well as examining factors that may be important to climate change and carbon balances. Examination is made of not only ecologically important but also political factors that may have an impact upon the management and carbon dynamics of blanket bog in the UK. This chapter has written contributions by Neil Wilkie concerning the Peatland Management Scheme, LIFE Nature and Heritage Lottery sections, Mike Wood for the Scottish Forestry Grants Scheme, and Mandy Gloyer for Agri-environment Schemes and Land Management Contracts. 1.1 Introduction There is a prevailing awareness that changes in climate at the global scale are a direct consequence of human activity and are predicted to persist for decades even under the most optimistic scenarios (Hulme et al., 2002; Hulme, 2005; King, 2005). That these changes will have associated effects on biodiversity is also likely (Hulme, 2005; King, 2005). The ability to address losses in biodiversity and global climate change requires the scientific understanding of biogeochemical cycles and how the processes such as disturbance affect biotic survival. Untangling the interactions of human activity and their effects on biological processes are some of the most earnest and challenging research questions faced by ecologists, spanning local, national and global scales. Global climatic change is expected through the enhancement of the earth's natural greenhouse effect by the rising concentrations of certain atmospheric greenhouse gases. The natural greenhouse effect arises from absorption of outgoing infrared radiation by greenhouse gases which is then emitted in all directions including to the earth's surface keeping the surface at a higher temperature (~14 oC) than would be the case in the absence of this effect (IPCC, 2001). Carbon dioxide (CO2) is a powerful greenhouse gas and may contribute 60 % of observed global warming effect (Grace, 2004). The evidence that concentrations of CO2 have been rising in the atmosphere is unequivocal (IPCC, 2001), Figure 1.2 illustrates the rising concentrations recorded at Mauna Loa from 1958 until 2004.
1
carbon dioxide concentration (ppm)
Chapter 1
380 370
monthly mean smoothed trend
360 350 340 330 320 310 1958 1966
1974
1983
1991 1999 2004
Figure 1.2: Rising concentrations of CO2 recorded at Mauna Loa (Keeling & Whorf, 2005). The observed rising concentration is not the only perceptible phenomenon shown by Figure 1.2, there is also an important seasonal drawdown due to northern hemisphere vegetation photosynthesis emphasising the importance of the biotic factors in carbon cycle. The rise of CO2 in the atmosphere correlates with increases in fossil fuel consumption due to industrial activity (IPCC, 2001). There are several other gases that contribute to the overall greenhouse effect these include direct greenhouse gases, such as, methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulphur hexafluoride (SF6) and indirect greenhouse gases, nitrogen oxides (NOx, as NO2) carbon monoxide (CO) non-methane volatile organic compounds (NMVOC) and sulphur dioxide (SO2) (IPCC, 2001; Baggott et al., 2004). Peatland ecosystems exchange both CO2 and CH4 but also represent a large store of carbon within the peat and host a distinctive assemblage of species, which, if lost, would decrease global biodiversity and potentially increase atmospheric carbon.
2
Chapter 1 Activities carried out on peatland ecosystems that may bring about these effects include drainage, agricultural improvement, burning, the effects of large herbivores, peat extraction and climate change. More than 85% of all peatlands are located in the northern temperate, boreal and arctic zones. These ecosystems (including tundra and boreal forests) are estimated to store 1.2 x 1018 g C (O'Neill, 2000). Bogs and fens alone account for approximately 3.0 – 4.6 x 1017 g C within an area of approximately 350 million hectares (O’Neill, 2000 and references therein). This may be equivalent to the total amount of carbon present in the atmosphere today (Clymo et al., 1998). With the uncertainties of ecosystem response to global climate change, the importance of conserving this carbon store cannot be overstated. The term peatland covers a wide range of peat-forming vegetation including tundra, boreal forests, fens and bogs (O'Neill, 2000), although the most important peatland habitat in the UK is blanket bog. Blanket bog can be defined as areas of semi-natural vegetation over-lying peat of at least 0.5 m depth and forming a blanket over moderately sloping ground (NCC, 1990). It is regarded as the most extensive seminatural land habitat in the UK covering at least 1.4 million hectares (Lindsay, 1995). The Flow Country in Sutherland and Caithness, in the north of Scotland, may be the largest area of continuous blanket bog in the World (Lindsay et al., 1988). The UK holds 10-15% of the total world area of this habitat (Lindsay, 1995) and, of this, Scotland holds over 1 million hectares; considering the UK is only approximately 0.16% of the global land mass this emphasises the importance of this habitat. The importance of the peat carbon store in the UK is demonstrated in Figure 1.1 whereby the majority of soil carbon can be seen to be located within Scotland and the majority of this constitutes blanket-peat. It is estimated that peatlands with a depth of over 45 cm contain 50% of all soil carbon and up to 40 times that which is contained within terrestrial vegetation in the UK (Cannell & Milne, 1995; Milne & Brown, 1997).
3
Chapter 1
Figure 1.2: The soil carbon content (kt km-2) of the United Kingdom (Milne & Brown, 1997) with the extent of blanket bog in Scotland (inset) defined as land with a depth of peat over 0.5 m (Lindsay, 1995). The development of blanket bog is a function of past and present environmental factors (e.g. climate, geology, geomorphology) and of the nature, intensity and history of human impact (Steiner, 1997). Bog ecosystems can be divided into two layers, the active growing layer (the acrotelm) and the layer of accumulated peat (the catotelm) (Ingram, 1978). Active blanket bog is an unbalanced system where plant production in the acrotelm exceeds the combined losses from decomposition of organic material and leaching of organic and inorganic carbon compounds (Vitt, 2000).
Gaseous carbon exchanges with the atmosphere are dominated by the
exchange of carbon dioxide (CO2) and methane (CH4). The net balance between the
4
Chapter 1 processes of photosynthesis and respiration determine the net gaseous CO2 exchange and establish whether the peatland is a sink or source of CO2. Gaseous exchange of CH4 is dominated by the process of methanogenesis in the catotelm emitting CH4 to the atmosphere. Oxidation of CH4 termed methanotrophy also occurs under aerobic conditions in the acrotelm but the balance between these two processes is generally in favour of methanogenesis, i.e. peatlands are a source of CH4. Both methanogenesis and methanotrophy are carried out by micro-organisms, generally bacteria. Other exchanges of carbon include the export of particulate and dissolved carbon into river systems and losses to the atmosphere through fire. The relative importance of these processes to the total carbon balance varies spatially and temporally.
The carbon balance of UK blanket peat is a major factor in
assessing UK’s greenhouse gas emissions (Milne & Brown, 1997). The majority of the blanket bog resource in the UK is subjected to management by agricultural drainage (moor grips), grazing, burning and forestry, but with the exception of forestry (Hargreaves et al., 2003), the variability of carbon dynamics under different types of management has yet to be quantified in the UK. The Scottish Biodiversity Strategy recognises the importance of peatlands in relation to climate change, but active blanket bog is not only important in terms of carbon, it is also classed as a priority habitat under the European Union Habitats Directive (source JNCC). The most important species in conservation terms are within the genus Sphagnum and those species associated with them. Active blanket bog is by definition a habitat that is actively accumulating peat, thus sequestering carbon. The conservation of active blanket bogs in the UK is focused on achieving the best representation of hydromorphological types, plant communities, and plant and animal species (JNCC, 1994). However, the active status of bogs in the UK is unknown, therefore, conserving diversity without information on carbon dynamics may not ensure that designated sites are actively sequestering carbon. However, in addition to conserving key species and habitats, it may be possible to manage bogs for carbon sequestration, or more likely the minimisation of carbon losses and the conservation of biodiversity as well as responding to climate change. Currently
5
Chapter 1 research is still required for the realisation of the key objectives and targets of UK biodiversity commitments, which include the encouragement of appropriate grazing, burning and other management practices on blanket bog habitats, as well as the restoration of degraded blanket bog to favourable condition by 2015 (Haines-Young et al., 2000). A better understanding of the impact of policy initiatives (e.g. agri-environment schemes, Scottish Natural Heritage Peatland Management Scheme) on carbon flux of blanket bogs is also required. Such findings are essential to provide information for Government reporting on greenhouse gas emissions, research on ecosystem response to climate change and reviews of Government-funded land-management schemes. 1.2 Geographical Extent of Blanket Bog and Management Relevant to Scotland The blanket bog resource in the UK is subjected to a variety of different management practices, these include drainage, grazing, burning, ecological restoration and forestry. As the effects of forestry have been investigated elsewhere (Hargreaves et al., 2003) this section will focus mainly on the management practices of drainage, grazing, burning and ecological restoration techniques. This section concentrates mainly on a Scottish perspective as much of the information on geographical extent has had more recent attention for Scotland and the majority of the UK blanket bog resource is in Scotland. 1.2.1 Total blanket bog resource In a recent review of climate change and organic soils in Scotland Chapman et al., (2001) noted with disappointment that after 50 years of peat survey there is still no definitive estimate for the geographical extent of peatlands in Scotland or the UK (Table 1.1). Part of the difficulties in reaching a reliable estimate for the geographical extent lie in defining the blanket bog habitat. This ultimately depends on the depth of peat chosen which has varied from a depth of over 1 m, to those peats over 0.3 m deep, and on the mapping technique. In Scotland there are extensive areas of vegetation that are essentially bog vegetation but overlie peat that is much shallower than 1 m, for example, the Lammermuir Hills, ignoring this is likely to underestimate
6
Chapter 1 the carbon store. Also, if pool systems are not taken into account then estimates would tend to be overestimates; it is unknown whether pool systems are considered in estimates of carbon storage. There are also areas that may once have been mapped as blanket bog but through persistent management practice are now classified as a different habitat e.g. Calluna moorland. These may still have a sizeable carbon store, though estimates based on vegetation cover may also underestimate the carbon store. It is also noticeable that the only estimate in Table 1.1 bounded by error estimates is that of the CS 2000 survey, this tends to imply a precision to estimates that is not actually evident. In a recent analysis using NVC survey data and comparing it with the SBBI and LCS 88 estimates, it was found that only 55.5% of the SBBI classifications were in agreement with the NVC whereas 69.7% of the LCS 88 classifications matched the NVC (Andrew Coupar pers comm. 2005). Assuming the NVC surveys themselves were accurate, this may suggest that the LCS 88 data gives a more accurate reflection of the extent of blanket bog (Andrew Coupar pers comm. 2005). There is also no agreement on the amount of carbon within these soils that is used to calculate the overall storage value. The reasons for this uncertainty are partly due to the uncertainty of extent but also to do with uncertainties surrounding the parameters chosen for calculation (Chapman et al., 2001), for example bulk density; see Chapman et al. (2001) for a fuller discussion. Chapman et al., (2001) conclude that between 2000 and 4500 Mt C is likely to be stored in Scottish peaty soils. There is therefore a requirement for method refinement to lead to better estimation of carbon content and geographical extent:
A practical and absolute definition of what should be included in any mapping project including an agreed minimum depth of peat for inclusion.
A mapping method that not only allows a definitive estimate of geographical extent but also gives an assessment of the errors associated with the estimate.
A better understanding of the range of depth and bulk density of blanket peats in Scotland.
These aspects are currently under review by the Organic Soils Modelling Project who have used a classification of two broad groups of organic soils (MISR, 1984):
7
Chapter 1 Organic – mineral Includes all soils with an organic surface horizon < 50 cm thick and an organic carbon content > 14% (25% OM) Organic Includes all soils with an organic surface horizon > 50 cm thick (peats) and an organic carbon content > 14% (25% OM). Most peats have organic carbon contents well in excess of this value Given the uncertainties associated with estimating the total peatland resource and its carbon store, estimates for the extent of management practices on blanket bog will be of similar low precision. Table 1.2 summarises current knowledge of the geographical extent of each of the management practices examined in this thesis. It should be remembered that these are estimates and are likely to be spatially variable and are not mutually exclusive. These estimates in some cases represent a best guess, others such as those from the LCS 88 or SBBI, may appear to have more precision but they are also not bounded by any error or estimation of variation. They should therefore be treated with caution.
8
Chapter 1 Table 1.1: Estimates of the peatland resource of Scotland (Chapman et al., 2001) with additions from The Scottish Blanket Bog Inventory, (SBBI) (Quarmby et al., 1999; Johnson & Morris, 2000c, a, b, d, 2001), Land Cover Scotland 1988 (LCS 88) Andrew Coupar pers comm. 2005, and Countryside Survey 2000 (Haines-Young et al., 2000). a Assuming 50% C, b probably an underestimate, c using the estimated C content of 114 kg C m-2, d peat soils > 1 m deep but may include some 0.3 - 0.5 m deep. Area of peatland kha
% of
Carbon
Total
store
Area
Mt C
821
11
600a
820
11
(Bather & Miller, 1991)
821
11
(Robertson & Jowsey, 1968)
821
11
(Jowsey, 1973)
765
9.9
(MISR, 1984)
699
(blanket peat)
66
(basin peat)
789
1000a
10.2
720
(blanket peat)
69
(basin peat)b
Reference
(Robertson, 1971)
(Birnie et al., 1991)
(approx.)
1742
22.6
1986c
(Cannell et al., 1993)
2625
30.9
16412
(Howard et al., 1995)
2625
30.9
4523
(Milne & Brown, 1997)
2564 (blanket peat) 61
(basin peat)
1332
17.2
(Anon, 1998)
1742
22.6
(Cannell et al., 1999)
1096d
14.2
(Patterson & Russell, 2000)
1056 (blanket) 40
(other)
9
Chapter 1 Table 1.1 continued Area of peatland kha
% of
Carbon
Total
store
Area
Mt C
Reference
1927 (blanket bog)
SBBI
660 (peatland as a single feature)
LCS 88
366 (mosaics, peatland as primary component) 1131 (mosaics, peatland as secondary component) 2038 (standard error 168)
CS 2000
2339 upper limit 1754 lower limit Table 1.2: Estimates for the geographical extent of management on blanket bog in Scotland. 1 SBBI (Quarmby et al., 1999; Johnson & Morris, 2000c, a, b, d, 2001), 2 Land Cover Scotland 1988, * this may be as high as 450,000 ha (W Towers pers. com. 2005, from re-calculation of LCS88 data), 3 JNCC a Assumed figure, b Includes all peat, not just blanket bog. Types of Management
Geographical Extent (ha)
% of total area
Total Blanket Bog Resource1
1,927,000
100
Grazed
1,927,000a
100
Burnt
???
???
Drained
???
???
Eroded2
200,000*
10
50,000
2.5
SAC3
220,847
11
SPA3
261,108b
13
SSSI 1
384,702
20
192,480b
10
11,800
0.6
Used for Peat Extraction2 Statutory Conservation
Ramsar3 Under Restoration
10
Chapter 1 1.2.2 Geographical Extent of Grazing It would seem reasonable to assume that the entire area of blanket bog in Scotland has historically (Shaw et al., 1996) and is presently subjected to grazing of one type or another. However what is unclear is the intensity of grazing to which different areas are subjected. Further variability is likely to be introduced from the type of animals grazing on these bogs different animals produce very different effects due to size and pressure of footprint, oral morphology and diet preference. Large herbivores affect peatland systems in several different ways, including defoliation, uprooting, trampling, defecation and urination. Each of these activities will have a different impact on the peatland system. 1.2.3 Geographical Extent of Burning The extent of burning on blanket peats is not known, but the practice is regionally variable (Hamilton et al., 1997). Although natural fire in Scotland is rare, most blanket peat dominated by either Calluna vulgaris or Molinia caerulea will be prone to fire, either as a management tool for sheep or grouse, or as accidental or malicious wildfire. Severe ground fires that ignite the peat are rare, but can occur in blanket peat and then cause very considerable damage with loss of carbon to the atmosphere and a complete change in ecosystem function (Maltby et al., 1990). 1.2.4 Geographical Extent of Drainage The extent of drainage of blanket peats is not known. Stewart and Lance (1983) (Coupar et al., 1997) state that government grants for drainage reached a peak of 80,000 ha per annum in the 1950’s and the mean in the 1960-70s was 20,000 ha per annum. 1.2.5 Geographical Extent of Erosion Based on LCS 88 there are approximately 200,000 ha of eroded blanket bog (Andrew Coupar pers. comm. 2005), but estimates vary and it may be as large as 450,000 ha (W. Towers pers. comm. 2005). The type of erosion will vary from large areas of eroding bog devoid of vegetation to micro-eroded areas from, for example, animal trampling and hagging; the extent of this variability is unknown.
11
Chapter 1 1.2.6 Geographical Extent of Peat Extraction Based on LCS 88 there are approximately 50,000 ha of bog under cutting or extraction, this is predominantly domestic cutting (Andrew Coupar pers comm. 2005). 1.2.7 Geographical Extent of Conservation Details on statutory designated sites are held by SNH or JNCC. Extent of sites not under statutory designation but still actively conserved, such as Local Nature Reserves (LNR) or Wildlife Sites has not been collated but will be held by Local Authorities or Wildlife Trusts. Approximately 221,000 ha (11%) has been estimated to be designated Special Area of Conservation (SAC). The extent designated as Special Sites of Scientific Interest (SSSI) is a little larger as some SAC’s are a core area within a SSSI or some SSSI’s haven’t been designated as SAC’s. Also, the Lewis Peatlands are an SAC not underpinned by an SSSI designation so this adds to the SAC total but not the SSSI total. 1.2.8 Geographical Extent of Restoration Approximately 1,800 ha of trees have been removed from blanket peat under the LIFE Peatlands Project in Caithness and Sutherland and this should rise to 2,400 ha by December 2006. Moor grips are currently being blocked over approximately 10,000 ha rising to 15,000 ha by December 2006 again under the LIFE Peatlands Project (Neil Wilkie pers com. 2005). The extent of blanket bog that could be practically restored is considered to be the majority of the total afforested area excluding only those areas under forestry near to the conclusion of the first rotation and those that are severely eroded which are considered beyond recovery (Andrew Coupar pers comm. 2005).
12
Chapter 1 1.3 Carbon Cycle of Blanket Bog Figure 1.3 illustrates a simplified representation of the carbon cycle of a blanket bog. The main input identified in Figure 1.3 is the uptake of carbon dioxide by the process of photosynthesis. Carbon outputs include carbon dioxide from respiration and aerobic
decomposition
and
methane
oxidation,
methane
from
microbial
decomposition, and particulate and dissolved organic carbon as well as dissolved inorganic carbon in water that runs off into river systems. The relative importance of the various components illustrated in Figure 1.3 has been examined in many studies. Emissions of CH4 accounted for 16% of the net ecosystem exchange of carbon in an oligotrophic boreal pine fen (Alm et al., 1997). However, net ecosystem exchange of CO2 was estimated to account for 99% of the carbon balance in some circumstances in a patterned boreal peatland (Waddington & Roulet, 2000). The variability of carbon flux is due not only to factors such as the climate and seasonal timing, but also the microhabitat topography, i.e. hummock, lawn or hollow, and importantly the position of the water table. Classical theory suggested that Sphagnum growth and peat accumulation in hollows was rapid while hummocks declined (von Post & Sernander, 1910). This has since been discredited and recent flux studies, have supported stratigraphic evidence that hollows can represent a net loss to the system whereas hummocks and lawns can accumulate carbon (Bubier et al., 1995; Waddington & Roulet, 2000). The long-term water table position is also related to the carbon balance of bogs in a complex manner. However, vegetation cover can be a useful indicator of carbon flux and bryophyte communities are good predictors of CH4 flux because the distribution of bryophytes is related to the longterm water table position (Bubier et al., 1995). The most important peat forming vegetation includes Sphagnum spp. and members of the Cyperaceae and Ericaceae. Variability exists in the contribution to peat formation between and within these groups.
13
Chapter 1
CH4
Atmosphere CO2
Vegetation
Photosynthesis
Photorespiration
Diffusion
Diffusion
Aerobic decomposition
CH4 Oxidation
Diffusion Run off
Acrotelm
CH4 plant transport
DOC leaching
Diffusion
CH4 ebullition
Catotelm
Litter production Root exudation
Organic Run off
DOC
Organic carbon
carbon
Anaerobic decomposition
Mineral Soil
Peat accumulation
Groundwater
Figure 1.3: Schematic representation of the peatland carbon cycle, red arrows = losses of carbon, green arrows = gains of carbon. 14
Chapter 1 1.3.1 Carbon Dioxide The exchange of carbon dioxide in peatland ecosystems mirrors that of most other terrestrial ecosystems; inputs are gained by photosynthetic activity of plants and micro-organisms. There may also be deposition of carbon in precipitation but this is unlikely to be a significant amount. Losses are accounted for by respiration, aerobic decomposition, and the oxidation of methane by anaerobic decomposition in the catotelm. The balance between these inputs and outputs determines whether the system is a sink for carbon dioxide or a source (whether the system is an overall sink is determined by taking into account all other forms of carbon). By convention a flow of carbon into the system has a negative value and sources are given a positive value and this convention is adopted throughout this thesis. Thus the balance for carbon dioxide can be represented by the following simple related equations: NEP = P - Rp – Rh NPP = P – Rp GPP = P Where NEP is Net Ecosystem Productivity (also called net ecosystem exchange, NEE), P is CO2 uptake by photosynthetic activity, Rp represents respiration by plants Rh respiration by heterotrophic organisms, NPP is Net Primary Productivity and GPP is Gross Primary Productivity (Grace, 2004). Upland productivity (NPP) in the UK has been measured historically by clipping experiments where vegetation was marked and after a defined period of time, removed, dried and weighed (Welch & Rawes, 1965; Clymo, 1970; Clymo & Reddaway, 1971; Forrest, 1971; Clymo & Reddaway, 1972; Forrest & Smith, 1975; Rawes & Hobbs, 1979; Rawes, 1981, 1983). There is now a growing amount of literature examining various aspects of these ecosystem productivity relationships in peatlands on a variety of scales from leaf to entire ecosystems, using a wider variety of techniques than just harvesting such as static and dynamic chambers or micro-meteorological methods like eddy-covariance, especially in North America and Scandinavia. Static (non-steady state) and dynamic (steady state) chambers differ in that static chambers do not have a gas flow system and enclose a headspace above vegetation or soil, fluxes are then calculated from
15
Chapter 1 changes in the original headspace gas concentration, dynamic chamber fluxes are calculated from the change in gas concentration of the gas flowing through the chamber from input to output. In the UK gaseous exchange research gained momentum during the TIGER programme (Oliver et al., 1998) but the majority of work in the UK on blanket bog has centred on quantifying fluxes of methane (See Chapter 2). One puzzling aspect of carbon flux research is the large range of units reported in the literature and these do not always explicitly state which chemical compound or element they relate to, for discussion of this see Appendix. 1.3.2 Methane Peatlands emit methane, as do all wetlands, as a by-product of microbial anaerobic decomposition. Recent suggestions of aerobic methane production by terrestrial plants remain controversial (Keppler et al., 2006) but if confirmed these emissions are likely to be dwarfed by several orders of magnitude by peatland emissions. There has been much research to date on the emissions of methane from northern wetlands including blanket bog. These indicate there are large temporal and spatial variations in CH4 emission rates that need to be taken into consideration. The following examination of the controls on methane emissions is largely from two reviews (Bubier & Moore, 1994; Joabsson et al., 1999). Depth of water table, soil temperature and vegetation type have been identified as controls of CH4 production and net CH4 emissions. Species differences in physiology and morphology make the effects of vascular plant functioning on net CH4 emissions difficult to predict. Correlations between environmental variables and CH4 emission have been established and variables are very strongly inter-related and often counteract each other. Estimates for emissions vary (Whalen & Reeburgh, 1992) suggest tussock tundra globally emits 42 +/- 26 Tg yr-1 but other studies (see Bubier and Moore, 1994, Joabsson et al., 1999, and references therein) estimate emissions in the region of 18-35 Tg yr-1. Sites largely similar in vegetation and topography display large differences
in
emissions
when
between-sample
differences
in
vegetation
classification and climate are taken into consideration. Water table is a strong predictor of CH4 flux therefore vegetation patterns may be useful in predicting CH4 flux but to date there has been no agreement of spatial scales or the system of
16
Chapter 1 vegetation classification in the different studies conducted. The solubility of CH4 is low (23-40 mg l-1 at 0-20 oC) therefore, CH4 can escape through diffusion bubble ebullition or transport through vascular plants through aerenchymatous tissue. Studies in rice paddy fields indicate that 90% of CH4 flux arises from the tillers of rice. Root transport of oxygen to the soil can impact on mechanisms of methane production and oxidation. This transport can reduce methanogenic bacterial activity but CH4 oxidation may be stimulated, as methanotrophic bacteria are O2 limited. However the net effect of roots may increase CH4 emissions as models suggest CH4 transport in soil is reduced without roots. The atmosphere constitutes a sink for CH4 thus a diffusion gradient exists. Increased CH4 oxidation would decrease this gradient but increases in organic substrate released by plants would increase methanogenesis and hence the gradient. Stomatal closure is partly effective in reducing emissions but emissions are still evident even when stomata are closed. However, this indicates that species composition is important to the control of CH4 emissions. Methanogenic bacteria use simple substrates and initially rely on other bacteria to break down complex organic molecules into simpler molecules. Positive correlations between net primary production (NPP) and CH4 emission have been used to suggest an association between new plant production and methanogenesis (Whiting & Chanton, 1993). However, a causal link seems unlikely since the two processes are separated spatially and temporally (in terms of the substrate from plant production reaching methanogens which would at least lead to a time lag) and it would seem more likely that both effects are related to temperature. Some work also suggests links between light intensity and emissions of methane (Lloyd et al., 1998) again this is correlative, light effects may be indirect and emissions may be more directly related to changes in temperature and stomatal conductance. 1.3.3 Peatland Carbon Fluxes to River Systems This section is intended as an introduction to peatland carbon exports to river systems and does not represent an exhaustive review. Carbon exports from peatland ecosystems to rivers are mainly composed of dissolved organic carbon (DOC), particulate organic carbon (POC), dissolved inorganic carbon (DIC), and dissolved CO2 and CH4 (Dawson et al., 2002). The amount of carbon that is transported
17
Chapter 1 annually is thought to be one or two orders of magnitude lower than the exchanges commonly found between vegetation and the atmosphere or between the atmosphere and oceans. DIC is composed of HCO3- ions and free dissolved CO2 associated with gaseous carbon dioxide via the carbonate equilibrium (Stum & Morgan, 1981, cited in Dawson et al., 2002).
Free CO2 outgases further downstream until reaching
equilibrium with the atmosphere and concentrations show diurnal and seasonal variation (Dawson et al., 2001).
Losses of free CO2 can also be attributed to
photosynthetic activity of aquatic plants and phytoplankton but quantifying this seems elusive at present (Dawson et al., 2001). The distinction between POC and DOC is based on size. POC ranges between 0.45 and 1.0 µm and DOC includes suspended particles below 0.45 µm (Dawson et al., 2002). Isotopic evidence points to the terrestrial origins of stream DOC and suggests that most may be of recent origin (post-AD 1955) (Palmer et al., 2001); in other words the majority of the DOC in streams is not produced there but transported from other systems. Dissolved organic matter (DOM) includes other compounds as well as those containing carbon; DOC is about 50% of DOM (Tipping et al., 1999). Our understanding of how organic matter is mineralized and partitioned into carbon dioxide, methane, and dissolved organic carbon is still lacking (Blodau, 2002). In Canada it has been estimated that between 2.4–5.6% of the peat carbon is mineralized annually (59 - l40 µg C g-1peat d-1) from floating peat islands in reservoirs (St Louis et al., 2003) the authors suggest that fluxes of CO2 and CH4 from peat could last 18–42 years from point of entry into the reservoir. However, the partitioning to different carbon products was not addressed. Intuitively the larger the organic pool in the catchment area the higher the DOC output, but this is also affected by stream physics such as discharge rate (Dawson et al., 2002). DOC can affect downstream aquatic net primary production (Carpenter and Pace 1997, in Pastor, et al., 2003), microbial production (Hobbie 1992, Wetzel 1992 cited in Pastor, et al., 2003) and other biogeochemical cycles (Driscoll et al. 1980, Hemond
18
Chapter 1 1980, Jackson and Hecky 1980, McKnight et al. 1985, Thurman 1985, Guildford et al. 1987, in Pastor, et al., 2003) and can also attenuate visible solar and UV-B radiation (Schindler et al. 1990, 1996, Scully and Lean 1994, Morris et al. 1995, Williamson et al. 1999, in Pastor, et al., 2003). Losses of DOC within the stream system can be attributed to biotic as well as abiotic sources such as biofilm respiration, adsorption to algae and mineral surfaces, particularly Fe and Al oxides, and hydroxides (Pastor et al., 2003). The composition of DOC is also important when considering fluxes to the atmosphere. Approximately 20% is low molecular weight compounds such as carbohydrates, amino acids, peptides, nucleic acids and carboxylic acids that represent a ready resource to the biota (Thomas 1997, in Dawson 2001). The remaining 80% tends to be phenolics and fulvic, humic and hydrophilic acids that represent more refractory compounds (Thurman 1985, in Dawson 2001). This suggests that the majority of the DOC resource is difficult to break down and may take a long time to reach the atmosphere in the form of gaseous emissions. Although there can be significant outputs from peatland systems as implied above, the overall effects of these outputs in terms of the impact upon greenhouse gas emissions are still unclear. A significant proportion of POC may be stored in the sediments and DOC is transported through the river system (Worrall et al., 2003b), presumably either being consumed in the stream or eventually reaching the ocean. There has been an observed trend of increasing DOC concentrations in river catchments in the UK over the last two decades (Monteith & Evans, 2000; Worrall et al., 2003a; Worrall et al., 2004a; Worrall & Burt, 2005). The reason for this remains elusive but is likely to be a combination of complex factors, for example climate change and its influence on microbial processes (Freeman et al., 2001a; Worrall et al., 2004a; Worrall et al., 2004b). There are at least three mechanisms whereby climatic change could affect the DOC budget of peatlands (Pastor et al., 2003):
increased temperatures could increase the production (through increased decay rates) and/or microbial consumption of DOC, thereby changing DOC concentrations in drainage water,
19
Chapter 1
changes in the position of the water-table level could change DOC concentrations as different portions of the peat profile become susceptible to aerobic and/or anaerobic decomposition regimes, and
changes in the water budget and discharge could control DOC export independently of any changes in DOC concentrations.
Attempts to model the increase in DOC in the UK in relation to temperature and water table have not been successful (Worrall et al., 2004a). It appears that it is difficult to achieve a model that is an adequate representation of the processes that it attempts to explore. Daulat and Clymo (1998) consider that reporting the relationship between methane and temperature by activation energies in an Arrhenius plot is misleading, since there are probably complex causes. It is worth noting that an Arrhenius approach was used in the model for production of DOC by Worrall et al., (2004a).
Perhaps DOC production needs to be considered as a more complex
process. Indeed, in a subsequent model the lack of complexity is acknowledged (Worrall & Burt, 2005). As no one process accounts for such a complex biological phenomenon as the production of DOC it should be expected that simplified models ultimately fail, but in their failure they can reveal issues that require clarification. Worrall, et al., (2004a) consider the ‘enzymatic latch’ (Freeman et al., 2001b) the most likely explanation of their results. This proposes that the absence of oxygen in peatland environments is responsible for the inhibition of the enzyme phenol oxidase (Freeman et al., 2001b). Phenol oxidase increases decomposition as the recalcitrant phenols are broken down which would happen when water tables are lowered (Worrall et al., 2004a). Freeman et al., (2001) reported a doubling of CO2 flux with a doubling of phenol oxidase. As noted above DOC comprises low molecular weight compounds such as carbohydrates, amino acids, peptides, nucleic acids and carboxylic acids, refractory phenolics, and fulvic, humic and hydrophilic acids (Thomas 1997 in Dawson 2001). Therefore an increase in phenol oxidase activity should lead to further breakdown of complex organic compounds that make up DOC and an increase in CO2 flux; not to a simple increase in the production of DOC as proposed by Worrall, et al., (2004a). However, perhaps the increase in phenol oxidase activity leads to a preferentially increased rate of breakdown of larger
20
Chapter 1 fragments of POC thus producing more DOC. This poses an interesting question; do substrates of similar composition but differing particle sizes decompose at different rates? St. Louis et al., (2003) found that rates of mineralization of peat pieces were not different from rates of mineralization of larger peat blocks in reservoirs but these were much larger fragments than the particle sizes of POC or DOC. If this relationship follows for POC and DOC decomposition, then both should decompose at the same rate thereby still leading to a reduction in total DOC due to faster decomposition rates under higher phenol oxidase activity. In a study on Great Dun Fell, England it was found, contrary to the theory of Worrall, et al., (2004a) that DOM production was in fact lower during the lower water tables of drought conditions (Scott et al., 1998) . Also, molecular changes in the composition of DOC were noted indicating that changes to the decomposition process were evident (Scott et al., 1998; Scott et al., 2001). Conversely clear responses to temperature were found in a lysimeter transplant experiment in relation to a peaty gley and DOC production in northern England (Tipping et al., 1999). This may be in part due to enchytraeid worms, as a positive response between temperature, DOC concentration and enchytraeid abundance has been found in the northern Pennines (Cole et al., 2002). DOC production is undoubtedly a very dynamic process with factors such as temperature, oxygen availability and moisture influencing chemical degradation, solute dissolution and microbial activity (Scott et al., 2001). In considering all the points above and in relation to the recent increase in DOC production in the UK, we need to ask whether future DOC production will continue to increase. If this does indeed happen will stream processes increase the conversion of DOC to gaseous emissions to the atmosphere, and if so will this increase continue indefinitely? It may be that riverine ecosystems have a kind of carrying capacity for DOC and inputs above this capacity would increase transfer of DOC from terrestrial systems to rivers and thereby to the ocean, but not necessarily increase losses to the atmosphere. The question is does this carrying capacity exist and if so what controls influence it, for example, biotic population sizes, availability of mineral substrates, or temperature? This could have important implications for modelling the contribution of this type of carbon export from peatland systems to atmospheric carbon budgets.
21
Chapter 1 This limited review of carbon export from peatlands to river systems has found no work investigating this hypothesis. Estimated outputs in temperate and boreal river systems have been reported to vary between 10 and 100 kg C ha-1 yr-1 (Hope et al., 1997) on the higher end of this range the river Halladale in a blanket bog catchment in Sutherland has been recorded with an output of 103.4 kg C ha-1 yr-1 (Hope et al., 1997) (see Appendix for tables of reported fluxes) . It is encouraging that most authors record information on the management of the catchment areas they are researching, (DOC outputs are tabluated in the Appendix). There remain though, some fundamental questions requiring research particularly involving the mechanisms of DOC production, the influence of climate and the transfer of DOC to the atmosphere. 1.4 UK Peatlands and the Greenhouse Gas Inventory (GGI) The UK ratified The United Nations Framework Convention on Climate Change (UNFCCC) in December 1993 which came into force in March 1994 (Baggott et al., 2004). Implicit in the convention is the development, publishing and regular updating of estimates of national emission inventories for greenhouse gases (GHGs). The UK publishes figures annually; the greenhouse gases reported are: Direct Greenhouse Gases
Carbon dioxide (CO2)
Methane (CH4)
Nitrous oxide (N2O)
Hydrofluorocarbons (HFCs)
Perfluorocarbons (PFCs)
Sulphur hexafluoride (SF6)
Indirect Greenhouse Gases
Nitrogen oxides (NOx, as NO2)
Carbon monoxide (CO)
Non-Methane Volatile Organic Compounds (NMVOC)
Sulphur dioxide (SO2)
22
Chapter 1 In the context of blanket bog the most significant of these gases are CO2, and CH4, since these gases are emitted and sequestered as part of biological processes. At present peatlands in the UK contribute to the inventory as part of the land use change and forestry category, appearing as the upland drainage and peat extraction for fuel and horticulture sections in the sub-category 'other'. The total emissions reported for land use change and forestry were approximately 2.5% of the UK total in 2002 and are declining gradually but this was attenuated by the estimated removal of nearly 11682 Gg of CO2 (~2% of total emissions) by uptake in photosynthesis of forests (changes in woody biomass stock) and agricultural crops (removals in sub-category 'other') (Baggott et al., 2004); see Table 1.3. Table 1.3: Summary of sources and sinks of greenhouse gases important to UK peatlands for the year 2002. Units are Gg CO2 equivalents, (Baggott et al., 2004). Greenhouse gas source sink
CO2
CO2
emission
removal
550965
CH4
N2O
-11682
2098.4
132
13585
-11682
1.1
0.1
Changes in woody biomass stock
-
-10582
-
-
Forest and Grassland Conversion
259
1.1
0.1
-
9937
Included
-
-
-
-
Total UK National Land use change and forestry total
CO2 emissions and removals from soils
elsewhere Other Total
3389
-1100
Scotland is important to this in the greenhouse inventory for two reasons. Firstly, by far the highest density (t ha-1) of carbon in the UK’s soils is found in Scotland, attributable to the extent of natural soils the majority of which are peatland (Table 1.4). Secondly, because the inventory takes account of drainage of upland peat soils due to afforestation, the majority of which occurs in Scotland (Table 1.5), this is counted as a source of CO2, therefore the majority of this emission originates in Scotland. Upland drainage and peat extraction account for over 60% of the subcategory 'other' (Table 1.6).
23
Chapter 1
Table 1.4: Average soil carbon density (t C ha-1) for different land cover in the UK (Baggott et al., 2004). The high carbon content of the natural category is due to the inclusion of blanket bog and other peaty soils. Region cover
England
Scotland Wales
N. Ireland
Natural
487
1048
305
551
Woodland
217
580
228
563
Farm (Arable)
153
156
93
151
Farm Pasture
170
192
200
178
Other
33
141
43
102
Table 1.5: Activity and Emission Factor Data for Upland Drainage Afforested Peat
Emission rate
Annual
(kha)
(t C ha-1a-1)
Loss (kt C)
England
20
2
40
Wales
10
2
20
Scotland
160
2
320
Northern Ireland
10
2
20
UK
200
2
400
Table 1.6: Breakdown of the contribution of upland drainage and peat extraction to subcategory 'other' (Table 4) adapted from Baggott, et al., (2004) units are Gg CO2 equivalents for the year 2002. Greenhouse source sink
gas CO2
% of emissions
emission
from other sub-
(Gg CO2)
category
Total Other
3389
100
Upland Drainage
1466.67
43.3
Peat extraction
682.92
20.2
24
Chapter 1
1.4.1 GGI and Unaccounted Emissions from Peatland At present most emissions from land-use change and forestry arise from the emissions of CO2 from soil, which includes the cultivation of mineral soils, liming of agricultural soils and drainage, although there is an acknowledge 60% uncertainty in the values for emissions from soils (Baggott et al., 2004). As already stated the majority of upland drainage occurs in Scotland. However, the inventory only includes drainage on upland soils due to afforestation, there is no account for drainage due to moor-gripping, which, although there is no quantitative estimate of the geographical extent, is widespread. This is probably due to the cessation of moor-gripping in recent times and land use change is only accounted for after 1990 (Baggott et al., 2004). However, if restoration of peatlands is taken up on a large sale then the consequences on carbon budget should be taken account of. Methane emissions from land-use change and forestry are accounted for entirely by the Forest and Grassland conversion category and arise from emissions from forests (Baggott et al., 2004). However, it has been suggested that under the terms of the Kyoto Protocol, peatlands could be used to meet reduction targets of carbon emissions under grazing land (Worrall et al., 2003b). If this is a realistic scenario it will be vital to take account of methane as well as carbon dioxide from peatlands. Further, restoration projects on blanket bog in Scotland affect both afforested and drained bogs and undoubtedly affect the carbon balance in the process. However, given the limited geographical extent of restoration projects at present this is unlikely to affect the greenhouse inventory greatly but if extended to larger areas information on the effects of these projects will be required to feed directly into the greenhouse inventory.
25
Chapter 1 1.5 Review of the effects of management on blanket bog and hypothetical effects on carbon fluxes This section summarises the effects of management on the biological components of blanket bog and then speculates what the effect of these may be on the carbon balance of the peatland. There are few direct measurements of the effect of management on carbon dynamics but see Garnett, (1998) and Garnett et al. (2000). 1.5.1 Grazing Shaw et al., (1996) reviewed the effects of grazing on blanket bog and wet heath and offer a more detailed examination of grazing than can be found here but much of what follows is adapted from that review. It seems likely that all areas of blanket bog in the UK are, or have been, subjected to grazing by both domestic and wild animals, and for the past 150-200 years rotational burning and grazing have been regularly practiced (Shaw et al., 1996). Grazing, like burning, may also be a contributory factor in the initiation of blanket peats in the UK (Shaw et al., 1996). Although it was not the primary objective of Shaw et al., (1996) to examine how management practices affect the carbon balance of blanket bog, their following points are worthy of note.
The effects of grazing vary according to stocking rates, wetness and condition of the site, type of grazing animal, time of year and length of time spent on the site.
Changes in vegetation composition and damage can be the result of trampling, which if severe, may result in bare ground. Effects can be localized, for example, around feeding points, fences, walls, etc. These areas are also affected by enrichment from dung and urine.
The effects of grazing on vegetation vary depending on the availability to herbivores of other habitats and food resources. There are interactions between management history and grazing that are difficult to separate, for example, grazing is often associated with burning.
26
Chapter 1 The immediate effect of grazing is a removal of vegetation and continual grazing can lead to a change in composition and structure such as the loss of heath to more Molinia and Eriophorum dominated vegetation (Shaw et al., 1996). Stocking rates are regionally variable and are partly dependent on the availability of other habitats for feeding. They can also be difficult to assess for example, Shaw et al., (1996) cite a ewe unit, which may include horses, and ewe counts which do not include lambs. Seasonality of use is also not reflected in stocking rates as a heavy winter and light summer use are evened out over the year (Shaw et al., 1996). Optimal stocking rates in terms of animal condition and with respect to vegetation would appear to be low but there is no definitive figure, this is dependent on location, site condition, climate, vegetation, etc. but is likely to be below 0.37 ewe ha-1 (Rawes & Hobbs, 1979; Shaw et al., 1996). The type of grazing animal can have an effect because of the different oral morphology and behaviour exhibited during the period on site. Cattle wrap their tongues round the vegetation and rip plants up; together with poaching this tends to produce a tussocky sward. Cattle are also less selective in diet preference than sheep (Shaw et al., 1996). Sheep bite and shear vegetation producing a much more even sward and are not as heavy. Breed and stock type can also show different effects as ewe and lambs are more selective in diet choice than wethers (Shaw et al., 1996). Goats have more of a preference for browsing woody vegetation. Horses and ponies tend to be less important in numbers on bogs but may be locally important for example on Exmoor and Shetland. Ponies tend to use the same site repeatedly for defecation leading to local areas of enrichment. Red deer are similar to sheep in their diet preference but proportionally eat less grass. Competition for the same areas may exacerbate damage but it is often difficult to separate the effects due to the different species (Shaw et al., 1996). Deer tend to prefer older rather than pioneer Calluna, grouse on the other hand require younger Calluna stems. Hares also favour pioneer Calluna and numbers can correlate with burning; like deer they can prevent regeneration of trees and in some cases Calluna (Shaw et al., 1996). Voles tend to
27
Chapter 1 prefer sites with Juncus and Molinia (Shaw et al., 1996) though not strictly ‘classic’ blanket bog vegetation these species can be prevalent on modified bog on deep peat. Seasonality of grazing also affects the disturbance to the site and vegetation, for example, Calluna tends to be eaten more in winter when grasses are less available this can lead to susceptibility to winter browning (wind and frost damage) if grazing is very heavy (Shaw et al., 1996). Before moving on to other management practices it is useful to examine the interactions of management practices. Thompson et al., (1995) present a simplified vegetation succession diagram, reproduced in Figure 1.4, useful in general terms for assessing interactions between burning, grazing and water table alteration. Although simplified this model is useful for examining relationships, however not all of it is based on evidence and some of the transitions are assumed (Shaw et al., 1996).
Figure 1.4: Simplified successional changes between bog and heath communities as affected by burning grazing and water table alteration (re-drawn from Thompson et al., 1995, cited in Shaw et al., 1996).
28
Chapter 1 1.5.2 Hypothesized Effect of Grazing on the Carbon Balance of Blanket Bog Grazing affects the carbon balance through effects on the vegetation (Figure 1.4), physical and chemical environment. These effects will vary according to the time spent in the habitat, season, animal species, breed, sex, age, and associated behaviour related to feeding, defecation, urination, travel and shelter. 1. Vegetation The physical acts of feeding and trampling can lead to the altering of vegetation composition structure and may lead to complete destruction of vegetation thereby leading to areas of bare peat. Altered vegetation composition will affect both photosynthesis and photorespiration altering the inputs and outputs of carbon dioxide to and from the atmosphere. A different vegetation composition would lead to an alteration of microbial decomposition in an unknown manner because of the different physical structure and chemical composition of different species. To date, the quantification of the effects of grazing on carbon fluxes has not been explored fully. 2. Physical Physical changes to the peat may come about through compaction by trampling and in extreme cases lead to erosion, see below. Through the actions of trampling and feeding the structure of the vegetation is also altered, thus altering carbon dynamics. 3. Chemical The nutrient balance of the bog can be altered by defecation, urination and the removal of plant matter. Further the removal of animals for slaughter or to other areas outside the blanket bog essentially translates to a removal of nutrients from the system. As carbon based life-forms this inevitably involves the transport of carbon out of the blanket bog system. Rawes and Heal (1978) (in Shaw et al., 1996) consider that there is little or no net income or loss from the bog in terms of N, P, K and Ca, but this work was conducted in the Pennines, which may be atypical in comparison to other areas of blanket bog in the UK in terms of utilisation for
29
Chapter 1 livestock. This carbon loss from the system will be transferred to the atmosphere in the short term as it is processed into food. 1.5.3 Burning Shaw et al., (1996) and Tucker (2003) reviewed the effects of burning on blanket bog and wet heath and more detailed examination can be found in these references. Although it was not the primary objective of these reviews to examine how management practices affect the carbon balance of blanket bog, the majority of what follows is summarized from these reviews. The following summary points from Shaw et al., (1996) are of note.
Most of the work to date investigating burning as a management tool has been conducted on grouse moors or lowland heaths, and so relates to a drier type of habitat than blanket bog.
Burning has physical, chemical and biological effects. The effects of fire are dependant on the vegetation, intensity and frequency of the fire, timing of the burn and the wetness of the habitat. Summer fires are likely to be most damaging for wildlife interest.
There will be indirect effects through changes in the physical habitat characteristics, plant species composition and vegetation structure and consequently microclimate.
Tucker (2003) summarised the impact of fire on selected upland species and the impacts on those species more prevalent in blanket bog are reproduced in Table 1.7 and Table 1.8. For a simplified model of how fire affects vegetation see Figure 1.3 above. Burning has been used for centuries and some authors believe that anthropogenic fire may have been responsible for he initiation of blanket bog in some areas (Moore et al., 1984), certainly evidence stretches as far back as Mesolithic period (Shaw et al., 1996). The intensity of fires varies according to the temperature reached and the
30
Chapter 1 speed. In extreme cases intense fires can ignite the peat removing the vegetation, produce a hard bitumen surface that can lead to increased runoff increased exposure increased heat and evaporation and an increased amplitude of temperature fluctuation decreased soil organic matter and nutrients, and seed bank destruction therefore making it difficult for plants to establish and may lead to erosion. This type of fire is more likely when ignition is accidental or malicious (Maltby et al., 1990; Legg et al., 1992; Tucker, 2003). The goal of managed fire is to remove and regenerate vegetation to improve food quality and vegetation structure, for example, Calluna for red grouse or grass and sedges for the ‘early bite’ (Shaw et al., 1996; Hamilton et al., 1997; Hamilton, 2000; Tucker, 2003). This latter strategy is used particularly on blanket bog in the north west of Scotland (Hamilton et al., 1997; Hamilton, 2000). Guidance on the use of fire is contained in the Muirburn Code (Anon, 2001), generally the burning of blanket bog is not recommended because of the detrimental effect it can have on the characteristic species and the risk of peat ignition, except where Calluna constitutes more than 75% of the vegetation (Anon, 2001) but these should be on long rotations (Shaw et al., 1996, Tucker, 2003, and references therein). However, Sphagnum species are not as sensitive as perhaps is assumed and do not always do badly under fire management (Hamilton, 2000; Tucker, 2003). There can also be interactions between fire and drainage because the water level can influence the effects of the fire as moist peat is insulated and severe burning can lead to increased peak flows in drainage ditches (Shaw et al., 1996). In concluding, Shaw et al., (1996) state that when burning (and grazing) are carried out indiscriminately these management practices are likely to be damaging to the wildlife interests of blanket bog and may even lead to loss of habitat. However, if conducted sensitively, both burning and grazing can have beneficial effects to some species of these habitats (though not all).
31
Chapter 1
Table 1.7: Summary of impacts of burning management on selected blanket bog species, species groups and blanket bog habitat based on Tucker (2003) and Hobbs et al., (1984). Species
Perennating organ & fire
Impacts
survival mechanism Calluna
Stem bases, protected by
Regenerates relatively rapidly after typical management fires, if burnt before the late
vulgaris
litter and persistent seed
mature phase. Re-establishes by seed from abundant long-lived seedbank if old stands are
bank
burnt or if hot fires damage basal stems. But seedling establishment is slow and may allow invasion by rhizomatous species. May not re-establish if burning is too frequent. Generally increases in abundance with long burning rotations (e.g. > 15 years) on bogs.
Empetrum
Buried branches
dominance in heathlands until overtopped by Calluna.
nigrum Erica tetralix
May be susceptible to fires but if prostrate stems are not destroyed then may gain temporary
Stem bases, protected by
Similar to Calluna, but favoured by shorter burning rotations of 6-10 years. May also be
litter and persistent seed
able to regenerate better in wetter habitats because its semi-prostrate lower branches are
bank
protected by Sphagnum and litter layers.
32
Chapter 1
Table 1.7 continued Species
Perennating organ & fire
Impacts
survival mechanism Eriophorum
Rhizomes
Often benefits from periodic fires, as can rapidly recolonise burnt areas from rhizomes, but is later out competed. May not survive post-fire conditions if significant changes in
angustifolium
moisture and pH. Eriophorum
Tiller apices within leaf
Rapidly regenerates after fire and probably resistant to hot fires due to tussocky growth
vaginatum
sheaves
form. Temporarily dominates after fires in blanket bogs and can remain dominant if burning rotations are less than 10 years.
Molinia
Tiller apices within leaf
Can regenerate rapidly after fire and often dominates (sometimes with E. vaginatum) under
caerulea
sheaves
frequent burning regimes.
Sphagnum
-
Often thought to be fire sensitive, but little evidence for this. Wet conditions may protect
mosses
species from fires and some can regenerate from deep buried fragments. Most impacts probably from peat damage and trampling, or due to exposure to drying or algal growth after removal of vegetation cover.
33
Chapter 1
Table 1.8 Extent of the practice of burning and advantages and disadvantages of this type of management on the blanket bog habitat (Tucker, 2003) Habitat
Extent of burning
Advantages
Disadvantages
Blanket Bog
Majority under some sort of
Eriophorum favoured may benefit Potential loss of fire sensitive species; can become
burning regime
black grouse and large heath butterfly dominated by Eriophorum on short rotations, or if abundance low. Some carefully Calluna on long rotations. Nutrient loss may be selected controlled burning may be significant Reduced peat formation and significant necessary to reduce fuel loads and risk risk of erosion and combustion of peat. Peat of wild fire
combustion and drying causes significant losses of carbon.
Increased
Eriophorum
increased methane flux.
34
may
cause
Chapter 1 1.5.4 Hypothesized Effect of Burning on the Carbon Balance of Blanket Bog Again as with grazing there are effects on the vegetation, physical and chemical environments, which will depend on the frequency intensity of the fire. An additional loss of carbon and other chemicals through fire will be to the atmosphere in smoke and ash. The consequences of fire on carbon balance will also be scale dependent. While the immediate consequences of fire are the loss of carbon to the atmosphere and death of important peat-forming species such as Sphagnum, in the intermediate term, the removal of shrub cover and litter may permit rapid recovery and expansion of Sphagnum and peat formation. In the long term, fire may promote increased Calluna dominance and changes to the hydrology of the bog that result in desiccation and oxidation of peat (Hamilton, 2000). There is also evidence that the perturbation of fire stimulates microbial activity within peat and probably increases the rate of decomposition (Maltby et al., 1990). Rates of peat accumulation have also been noted to be lower in areas that are burnt (Kuhry, 1994; Garnett, 1998; Garnett et al., 2000) suggesting that in terms of carbon sequestration burning may not be beneficial. Severe fire can lead to the direct combustion of peat and may lead to erosion thus exacerbating the carbon loss (Tucker, 2003). However the long-term impacts of burning are more complex than it would first appear as Calluna accumulates more carbon in the building and mature phases (Tucker, 2003). The removal of a dense shrub canopy has also been observed to benefit the recovery of Sphagnum species in some bogs (personal observation) this may be brought about by fire or other mechanical means with unknown implications for the carbon balance. 1. Vegetation The removal of vegetation through burning alters vegetation structure and competitive interactions between species thus leading to altered species composition. An increase in Eriophorum may cause increased methane flux to the atmosphere. Although burning is a different process, both burning and grazing affect the vegetation composition and structure and will therefore alter the carbon related processes of photosynthesis and respiration. To date, the quantification of the effects of burning on carbon fluxes on blanket bog has not yet been explored fully.
35
Chapter 1 2. Physical Physical changes to the peat depend on the frequency and intensity of the fire these are most likely to be extreme when associated with accidental fires with a high fuel load and that may lead to erosion and thus a loss of carbon, see below. 3. Hydrological Burning reduces the water storage capacity of the peat and, again, in extreme conditions may lead to areas of bare peat; these may increase evaporation and runoff which are likely to increase fluxes of DOC, etc. from the peatland system. The alteration of storage capacity may also lead to an altered water table thereby altering the balance between aerobic and anaerobic decomposition with consequences for the carbon balance. 4. Chemical Burning causes a short-term availability of nutrients and alteration to pH but there are undoubted losses from the system including carbon. Even though there is replacement from atmospheric inputs, there may be long-term shortfalls in the replacement of N, P and K (Tucker, 2003). The implications for this on the carbon dynamics are unknown at present most studies are limited in that they are concerned with short-term rather than long-term impacts (Tucker, 2003), but there will be impacts upon biological processes from the changing of nutrient availability. This will be further complicated by increased deposition of chemicals in upland areas from industrial pollution. 1.5.5 Drainage The practice of moor gripping on blanket peats has been continued for a number of centuries. Original drains were cut by hand but in more modern times by machine. Drains vary in size and depth. Drains can range from single drains for boundary demarcation to extensive herring bone patterns of moor grips 40-50 cm deep (see Figure 1.5). The desired effects of drainage are a lowering of the water table thus
36
Chapter 1 leading to an altered vegetation and more desirable area for sporting and agricultural activities (Coupar et al., 1997).
Figure 1.5: Artificial moor-gripping network on blanket peat near Forsinard, Sutherland in Scotland. Grid squares are 1 km. Wheeler & Shaw (1995) examined drainage effects on both raised mires and blanket peat. These systems are ecologically, if not morphologically, similar so their findings are still appropriate. Therefore, much of the following is taken from Wheeler & Shaw (1995). As drainage lowers the water table there can be an accelerated decomposition of the peat, a change in the physical properties of the peat and thus the hydrology, morphology and the ecology of the peatland ecosystem are altered (Wheeler & Shaw, 1995). Typical effects are increased subsidence, bulk density and amplitude of water table fluctuation with decreases in active porosity, water content,
37
Chapter 1 water storage coefficient and permeability (Wheeler & Shaw, 1995). The results are primary consolidation followed by shrinkage, secondary compression and finally wastage of the upper layers of the bog (Hobbs, 1986, cited in Wheeler & Shaw, 1995). The chemistry of the peat can be altered by the induction of biochemical oxidation mineralization and the release of H+ and nutrients altering the pH. In a damaged site frequent and long periods of drought may accentuate these processes leading to a sub-optimal pH for Sphagnum growth. The permanence of these effects is not known. On the vegetation, sustained lowering of the water table leads to a rise in Calluna vulgaris and Molinia caerulea and may result in invasion of birch, Betula spp. Longsustained lowering of the water table can lead to loss of typical bog species and loss of the acrotelm itself leading to the aeration of the catotelm, a faster decomposition of catotelmic peat and the cessation of peat accumulation and bog growth. Vegetation effects can take a long time to become evident and in one study were confined to the downslope side of the drains (Stewart & Lance, 1991). Stewart and Lance (1991) also found that cover of species dependent on high water tables had lower cover nearer to drains, cover of Calluna peaked after approximately 8 years and declines in Sphagnum were localized and took nearly 20 years to achieve statistical significance. The low hydraulic conductivity of the catotelm means that the effects of any one ditch are usually restricted to with a few metres either side of the ditch (Stewart & Lance, 1991). This is evident from the need to space ditches 10-20 m apart to provide sufficient drainage for the peat extraction (Wheeler & Shaw, 1995). Drainage will undoubtedly lead to a faster runoff in the immediate vicinity of the drain and longestablished drains can frequently be seen to have caused lowering of the peat surface for 5-10 m creating a parabolic peat surface and thus changing the hydrology of the bog.
38
Chapter 1 1.5.6 Hypothesized Effect of Drainage on the Carbon Balance of Blanket Bog As illustrated above, the effects of drainage act on the hydrology, vegetation and physical characteristics of the peatland. None of these effects act in isolation and are likely to interact with one another. The carbon balance is likely to be affected thus: 1. Hydrology a. Increased run off leading to increased exports of carbon to river environments. There is some work to support this hypothesis (Yeo, 1998). b. Lowered water tables leading to altered exchanges of gaseous carbon through the altered decomposition processes. This may lead to lowered methane emission but increased carbon dioxide emission. The long term dynamics of this have not been explored. 2. Vegetation As above in section on fire and grazing. 3. Physical Physical effects compound both hydrological and vegetation effects and thus are likely to compound effects on the carbon balance. 1.5.7 Erosion The most comprehensive studies of erosion have taken place in the Pennines of England by John Tallis. Although every situation could be regarded as unique and it could be argued that the Pennines may be atypical of blanket peat in the UK, the true value of these studies is that they have identified local erosion processes that have a wider applicability. Identifiable changes associated with erosion include: reduced species diversity, reduced Sphagnum cover, discontinuous plant cover and reduced productivity and peat accumulation (Tallis, 1997b). This led to the production of a simplified and generalized sequence of progression (see Figure 1.6 and Tallis, 1997b). Not all of these effects are displayed in any one eroded bog but this stresses the diversity of factors that are involved in the erosion process. Note the compound
39
Chapter 1 nature of many factors and the fact that many management practices are evident. Identified agents implicated in the erosion process are of both natural and anthropogenic origin including accidental fires of which there were 300 in the period 1970-1998 in the Peak District. At Holme Moss a particularly severe fire in 1700 is thought to be responsible for much bare peat today (Tallis, 1997b). Further agents include industrial pollution, sheep grazing, trampling, peat cutting and climatic impacts. These then may finally lead to erosion, which may take many forms from small areas of bare peat to fully formed integrated systems of gullies. Gully erosion is a feature of nearly all blanket peats in the UK and a mean erosion rate of 5.5 mm yr-1 has been postulated for the Peninne area indicating that a 1 m deep gully is approximately 200-250 years old with some of the deeper gullies being considerable older (Tallis, 1997b). Annual erosion rates in Shetland were 1- 4 cm yr-1 which may indicate that bare peat surfaces persist for 30-150 years for 1.5 m deep blanket peat, if erosion rates, geomorphological and management factors remain constant (Birnie, 1993). The evidence from the Pennines indicates that erosion is a long-term process. However, it may not be a permanent one: around 10% of the Moor House National Nature reserve was classified as re-vegetated former erosion (Garnett & Adamson, 1997).
40
Chapter 1
Figure 1.6: A simplified scheme of bog degradation and erosion redrawn from (Phillips, Yalden & Tallis, 1981, cited in Tallis 1998). There is quite a body of evidence on erosion and further work can be found in the following references (Tallis, 1959, 1964, 1965; Crisp, 1966; Stewart et al., 1966;
41
Chapter 1 Tallis, 1973, 1985, 1987; Bradshaw & McGee, 1988; Birnie & Hulme, 1990; Francis, 1990; Johnson et al., 1990; Stevenson et al., 1990; Birnie, 1993; Glenn et al., 1993; Heathwaite, 1993; Tallis, 1994; Tallis & Livett, 1994; Grieve et al., 1995; Tallis, 1995; Younger & McHugh, 1995; Fisher et al., 1996; Mackay & Tallis, 1996; Tallis, 1997a; Ellis & Tallis, 2000; Bragg & Tallis, 2001; Ellis & Tallis, 2001; Evans & Warburton, 2001; Wishart & Warburton, 2001; Campbell et al., 2002; McHugh et al., 2002; Waddington & McNeill, 2002; Ellis & Tallis, 2003; Warburton, 2003; Warburton et al., 2003; Warburton et al., 2004). 1.5.8 Hypothesized Effect of Erosion on the Carbon Balance of Blanket Bog The effects of erosion on the carbon balance are likely to be very similar to the effects of drainage see above, only sometimes on a much larger scale. The carbon balance is likely to be affected as above: 1. Hydrology, as in drainage. 2. Vegetation as in grazing and burning sections. 3. Physical as in drainage section. 1.5.9 Peat Extraction The effect of peat extraction depends entirely on the method and scale of extraction, which dictates the degree of severity to the peatland system. As noted above the majority of peat extraction in Scotland is done by domestic cutting for fuel. 1.5.10 Effect of Peat Extraction on the Carbon Balance of Blanket Bog The use of peat as a fuel in terms of carbon is similar to other fossil fuels in that it is an unsustainable resource and emits carbon to the atmosphere. The loss of carbon due to be peat extraction was calculated as 682.92 Gg CO2 for the year 2002 in the UK (Baggott et al., 2004). However this figure includes raised bogs and therefore the real total for blanket bog in would be less. 1.5.11 Conservation Conservation is not strictly a management practice and the effect of designation of a site may be to introduce or cease different management practices for the achievement
42
Chapter 1 of the conservation goals. Conservation therefore can be assessed by reference to the different management practices presented here. In the current context the management practices of interest are those that influence the carbon balance by conserving an active bog ecosystem, i.e. a bog accumulating carbon. 1.5.12 Hypothesized Effect of Conservation on the Carbon Balance of Blanket Bog The key here is to identify whether management by conservation is effective in conserving active status of bogs. This has yet to be assessed although SNH hold data on habitat condition for some if not all blanket bog SSSIs. 1.5.13 Restoration Restoration could be examined under conservation but is detailed separately here because it is a fairly new practice and also due to the extensive restoration projects currently undertaken in the Caithness and Sutherland Peatlands area, where much of the field work for this thesis was conducted. Before examining the effect of restoration we need to define what blanket bog ecological restoration is, accordingly I define this as: ecological restoration of blanket bog is defined as any management practice that is deliberately undertaken to restore ecological processes, communities and/or species to semi-natural condition, thereby enhancing the ecosystem of blanket bog. Though broad, within this definition there are some key elements that need elaboration. A management practice that is deliberately undertaken must have the enhancement of the blanket bog ecosystem as an objective target. Further it is explicitly acknowledging that anthropogenic influence is required to achieve seminatural status and it is also required to maintain that status (unlike definitions of naturalness for woodlands which imply naturalness without the influence of people (Peterken, 1996)). By stating that the objective is 'semi-natural condition' there is no implication that these practices can ‘turn back the clock’ and deliver any particular ecosystem that was present in the past. It is unfortunate then that the word restoration
43
Chapter 1 has entered into common use for conservation projects, remediation or rehabilitation may be better but since restoration is commonly used it is retained here. Finally, by including ecological processes, communities and species, this definition recognises that none of these entities exists in isolation and that they are all are required to enhance a blanket bog ecosystem. Recent restoration in Caithness and Sutherland has concentrated on two distinct types of degraded mire: those planted with trees and those affected by moor-grips in Caithness and Sutherland. There are therefore different methods for tree removal and drain blocking, although in reality drains also need to be blocked after tree removal. Although the objectives of restoration include promotion of birds and invertebrates, the main effects examined here will be those on vegetation and hydrological impacts that primarily affect the carbon balance. Tree Removal: In Caithness and Sutherland where the majority of the restoration projects have been carried out three methods of tree removal have been used chainsaw felling, mulcher, and mechanical tree snipper. Either, the trees are felled to waste leaving mulch behind or with cut trees and brash placed into furrows, helping to impede drainage, or the trees may be removed for use in bio-fuel or other commercial/community uses. Drain blocking: The objective of damming drains is to raise water tables and in so doing help regulate base and peak flow rates to the respective burns, with a concomitant reduction in the frequency of spates. Outcomes are the restoration of peatland vegetation as well as hydrology. Drains are dammed, for example at 20 to 25 cm drops, either using a mixture of materials such as plastic pile and peat dams constructed by hand or using a low ground pressure digger, depending on the situation. 1.5.14 Hypothesized Effect of Restoration on the Carbon Balance of Blanket Bog The removal of trees may result in a reduction in evapo-transpiration and hence raise the water table and promote the recovery of vegetation. The quality of the woody
44
Chapter 1 matter left to decay may also be important in determining the rates of decay and carbon release or retention in the newly restored acrotelm. The effects of blocking drains will be to alter the dynamics of the carbon flux in favour of methane production. As stated above it is the balance between the methane emitted and carbon dioxide fixed that is important in determining if the bog is a source or sink to the atmosphere. It remains to be seen whether the methane pulse shown by many studies (Anderson, 2001) is a transient phase and lessens over time as the vegetation becomes more established. 1.6 Examination of Policy Mechanisms for Blanket Bog Restoration Although restoration is relatively new there is likely to be increase in restoration projects especially if policy mechanisms are used as encouragement. There is included here then a short appraisal of current policy mechanisms available to landowners that have a direct or indirect link to peatland restoration. There are several mechanisms in place that have been used for blanket bog restoration. It also is likely that future policy mechanisms could be used for the financing of blanket bog restoration not only for specific conservation projects but also for integrated projects within the rural farming environment. The restoration of blanket bog in the UK is a relatively recent phenomenon. It has yet to be considered on a large geographical scale. Part of the problem may be a lack of awareness of what options are available to landowners with regard to peatland restoration and the complexity of the granting system Peatland Management Scheme (PMS - administered by Scottish Natural Heritage [SNH]): This is a scheme for SSSI landholders that include options for 'peatland restoration'. Although a very successful scheme in terms of uptake, only a few landholders have done any restoration work through this scheme. The current LIFE Peatlands Project (LPP) is trying to promote a larger uptake of this aspect of the PMS by specifying that SNH will do five restoration schemes as part of the project. The added resources of the LPP are helping SNH progress on this. All restoration work carried out to date under this scheme has been blocking hill drains.
45
Chapter 1 Scottish Forestry Grants Scheme (SFGS – administered by Forestry Commission Scotland [FCS]): In June 2003 this replaced the Woodland Grant Scheme (WGS) 'Woodland Improvement Grant' that included aspects of tree removal from deep peatland. SFGS, like its predecessor, is still a restructuring grant for ‘improving woodland biodiversity’ with funded open-ground restoration limited to 20% of the forest area (FCS, 2003, 2005). In practice, WGS & SFGS grants are for small scale (mostly 10's of hectares) tree removal from, for example, the edge of a Natura site. In February 2005 FCS increased the rate of SFGS grant paid for open ground restoration by tree removal to 90% of ‘standard costs’, but the limitation of only 20% open-ground restoration of the forest remains. Going beyond 20% of forest area removed under SFGS may require a cultural shift and change in FCS’s remit, with closer linkage with SEERAD/SNH and their support to manage the restored peatland. FCS may feel constrained by being the Forestry rather than ‘Peatland’ Commission. LIFE Nature (EU): This has been the only funding source that has allowed landscape scale restoration work in peatlands. Favourable LIFE applications focus directly on the 'threats' to a Natura site. In the case of the Caithness and Sutherland peatlands these were identified as mainly hill drains and forestry. 2005 is the last year for applications for LIFE Nature projects - future funding of Natura work will come through the Rural Development Regulation 2007-2013. Heritage Lottery Funding: The RSPB has had some success in acquiring conifer plantations on peatland areas for restoration purposes including the felling of conifer trees. Agri-environment Schemes: These schemes have been in operation in Scotland since 1987. They are designed to encourage farmers and crofters to manage their land for the benefit of Scotland's wildlife and habitats. Participation in the schemes is for a minimum of five years. In benefiting wildlife and habitats there may also be a pay off in terms of carbon budgets particularly in the case of blanket bog where a well-
46
Chapter 1 managed functioning peatland is more likely to be a carbon sink than a carbon source. There are certain schemes that are no longer open for applications such as the Environmentally Sensitive Area (ESA) and Habitats Scheme, as such these schemes would not cover future peatland projects but existing agreements may still be benefiting peatland areas. The Rural Stewardship Scheme (RSS): RSS is part of the Scottish Rural Development Plan. It replaced the Countryside Premium Scheme (CPS) and provides assistance to landowners and managers for the adoption of environmentally friendly practices and to maintain and enhance particular habitats and landscape features. The Moorland Management Option for RSS would cover elements of peatland restoration. Organic Farming: This Scheme, which is part of the Scottish Agri-Environment Programme, came into operation in July 1994. It provides assistance to farmers and crofters who wish to convert to organic production. Although no direct payments would be made through this scheme for peatland restoration the practice of organic farming in upland areas may have an indirect benefit to peatland environments. Land Management Contracts: The Land Management Contract (LMC) Menu Scheme was launched on 25th February 2005. The scheme for 2005 contains an option for management of moorland grazing, which aims to benefit a diverse range of habitats of conservation interest within moorland. The Menu Scheme is lower level than RSS, and does not contain prescriptions for enhancement through management. The future development of LMCs may provide further opportunities to include other aspects of peatland restoration, and the full LMC model, due to be launched in 2007, could contain further prescriptions targeted at these.
47
Chapter 1 1.7 Conclusions
Peatlands are large carbon stores.
The largest peatland habitat in the UK is blanket bog.
Blanket bog in the UK is subjected to varying types of management including grazing, burning.
The geographical status of blanket bog in the UK is at present equivocal including the extent of management practices.
Peatlands have like many other ecosystems a complex carbon cycle involving exchanges of CO2, CH4 and exports of organic carbon into river systems.
Conservation of the carbon stored and carbon exchange processes of blanket bog peatlands habitat is vital for the consideration of the greenhouse gas balance of the UK.
At present, certain peatlands may either be sinks or sources of carbon but more research is required particularly in the UK.
Restoration of blanket bog is a relatively recent practice
Quantification of the carbon dynamics of the UK blanket peat taking into account different vegetation composition and management regimes may reveal opportunities for the restoration of ecological processes, but whether or not peatlands can be turned into carbon sinks by ecological restoration remains to be answered.
The only way to allow blanket bog ecosystems to adapt to climate change may be through the restoration of ecological processes.
1.8 References Alm, J., Talanov, A., Saarnio, S., Silvola, J., Ikkonen, E., Aaltonen, H., Nykanen, H., & Martikainen, P.J. (1997) Reconstruction of the carbon balance for microsites in a boreal oligotrophic pine fen, Finland. Oecologia, 110, 423-431. Anderson, R. (2001). Deforesting and restoring peat bogs, a review, Rep. No. Technical Paper 32. Forestry Commission, Edinburgh. Anon (1998) Scottish Environment Statistics. The Stationary Office, Edinburgh.
48
Chapter 1 Anon (2001). The Muirburn Code. Scottish Executive, Edinburgh. Baggott, S., Brown, L., Milne, R., Murrells, T.P., Passant, N., & Watterson, J.D. (2004). UK Greenhouse Gas Inventory, 1990 to 2002: Annual Report for submission under the Framework Convention on Climate Change, Rep. No. Issue 1.2. Department for Environment, Food and Rural Affairs, AEA Technology Abingdon. Bather, D.M. & Miller, F.A. (1991). Peat Utilisation in the British Isles. Centre for Agricultural Strategy, Reading. Birnie, R.V. (1993) Erosion Rates on Bare Peat Surfaces in Shetland. Scottish Geographical Magazine, 109, 12-17. Birnie, R.V., Clayton, P., Griffiths, P., Hulme, P.D., Robertson, R.A., Soane, B.D., & Ward, S.A. (1991). Scottish Peat Resources and their Energy Potential, Rep. No. ETSU B 1204. Department of Energy. Birnie, R.V. & Hulme, P.D. (1990) Overgrazing of Peatland Vegetation in Shetland. Scottish Geographical Magazine, 106, 28-36. Blodau, C. (2002) Carbon cycling in peatlands - A review of processes and controls. Environmental Reviews, 10, 111-134. Bradshaw, R. & McGee, E. (1988) The extent and time-course of mountain blanket peat erosion in Ireland. New Phytologist, 108, 219-224. Bragg, O.M. & Tallis, J.H. (2001) The sensitivity of peat-covered upland landscapes. Catena, 42, 345-360. Bubier, J.L. & Moore, T.R. (1994) An ecological perspective on methane emissions from northern wetlands. Trends in Ecology and Evolution, 9, 460-464. Bubier, J.L., Moore, T.R., & Juggins, S. (1995) Predicting methane emissions from bryophyte distribution in northern Canadian peatlands. Ecology, 76, 677-693. Campbell, D.R., Lavoie, C., & Rochefort, L. (2002) Wind erosion and surface stability in abandoned milled peatlands. Canadian Journal of Soil Science, 82, 85-95.
49
Chapter 1 Cannell, M.G.R., Dewar, R.C., & Pyatt, D.G. (1993) Conifer plantations on drained peatlands in Britain: a net gain or loss of carbon. Forestry, 66, 353-369. Cannell, M.G.R. & Milne, R. (1995) Carbon pools and sequestration in forest ecosystems in Britain. Forestry, 68, 361-378. Cannell, M.G.R., Milne, R., Hargreaves, K.J., Brown, T.A.W., Cruickshank, M.M., Bradley, R.I., Spencer, T., Hope, D., Billett, M.F., Adger, W.N., & Subak, S. (1999) National inventories of terrestrial carbon sources and sinks: The UK experience. Climatic Change, 42, 505-530. Chapman, S.J., Towers, W., Williams, B.L., Coull, M.C., & Paterson, E. (2001). Review of the contribution to climate change of organic soils under different land uses. Scottish Executive Central Research Unit, Edinburgh. Clymo, R.S. (1970) The growth of Sphagnum: methods of measurement. Journal of Ecology, 58, 13-49. Clymo, R.S. & Reddaway, J.F. (1971) Productivity of Sphagnum (bog moss) and peat accumulation. Hidrobiologia, 12, 181-192. Clymo, R.S. & Reddaway, J.F. (1972). A tentative dry matter balance for the wet blanket bog on Burnt Hill Moor House NNR, Rep. No. Aspects of the Ecology of the Northern Pennines. Occasional Papers No. 3. Nature Conservancy. Clymo, R.S., Turunen, J., & Tolonen, K. (1998) Carbon accumulation in peatland. Oikos, 81, 368-388. Cole, L., Bardgett, R.D., Ineson, P., & Adamson, J.K. (2002) Relationships between enchytraeid worms (Oligochaeta), climate change, and the release of dissolved organic carbon from blanket peat in northern England. Soil Biology & Biochemistry, 34, 599-607. Coupar, A., Immirizi, C.P., & Reid, E. (1997) The nature and extent of degradation in Scottish blanket mires. In Blanket mire degradation causes, consequences and
50
Chapter 1 challenges (eds J.H. Tallis, R. Meade & P.D. Hulme), pp. 90-100. The Macaulay Land Use Research Institute, Aberdeen, University of Manchester. Crisp, D.T. (1966) Input and Output of Minerals for an Area of Pennine Moorland Importance of Precipitation Drainage Peat Erosion and Animals. Journal of Applied Ecology, 3, 327-&. Dawson, J.J.C., Billett, M.F., & Hope, D. (2001) Diurnal variations in the carbon chemistry of two acidic peatland streams in north-east Scotland. Freshwater Biology, 46, 1309-1322. Dawson, J.J.C., Billett, M.F., Neal, C., & Hill, S. (2002) A comparison of particulate, dissolved and gaseous carbon in two contrasting streams in the UK. Journal of Hydrology, 257, 226-246. Ellis, C.J. & Tallis, J.H. (2000) Climatic control of blanket mire development at Kentra Moss, north-west Scotland. Journal of Ecology, 88, 869-889. Ellis, C.J. & Tallis, J.H. (2001) Climatic control of peat erosion in a North Wales blanket mire. New Phytologist, 152, 313-324. Ellis, C.J. & Tallis, J.H. (2003) Ecology of Racomitrium lanuginosum in British blanket mire - evidence from the palaeoecological record. Journal of Bryology, 25, 715. Evans, M. & Warburton, J. (2001) Transport and dispersal of organic debris (peat blocks) in upland fluvial systems. Earth Surface Processes and Landforms, 26, 10871102. FCS (2003). Scottish Forestry Grants Scheme - Applicants' Booklet. Forestry Commission Scotland, Edinburgh. FCS (2005). SFGS Standard Costs & Specifications Booklet 2nd Edition, February 2005. Forestry Commission Scotland, Edinburgh.
51
Chapter 1 Fisher, A.S., Podniesinski, G.S., & Leopold, D.J. (1996) Effects of drainage ditches on vegetation patterns in abandoned agricultural peatlands in central New York. Wetlands, 16, 397-409. Forrest, G.I. (1971) Structure and production of north Pennine blanket bog vegetation. Journal of Ecology, 59, 453-479. Forrest, G.I. & Smith, R.A.H. (1975) The productivity of a range of blanket bog vegetation types in the northern Pennines. Journal of Ecology, 63, 173-202. Francis, I.S. (1990) Blanket Peat Erosion in a Mid-Wales Catchment During 2 Drought Years. Earth Surface Processes and Landforms, 15, 445-456. Freeman, C., Evans, C.D., Monteith, D.T., Reynolds, B., & Fenner, N. (2001a) Export of organic carbon from peat soils. Nature, 412, 785-785. Freeman, C., Ostle, N., & Kang, H. (2001b) An enzymatic 'latch' on a global carbon store. Nature, 409, 149. Garnett, M.H. (1998) Carbon storage in Pennine moorland and response to change. PhD Thesis, University of Newcastle-Upon-Tyne, Newcastle-Upon-Tyne. Garnett, M.H. & Adamson, J.K. (1997) Blanket mire monitoring and research at Moor House National Nature Reserve. In Blanket Mire Degradation, causes, consequences and challenges. MacAulay Land Use Research Institute, University of Manchester. Garnett, M.H., Ineson, P., & Stevenson, A.C. (2000) Effects of burning and grazing on carbon sequestration in a Pennine blanket bog, UK. The Holocene, 10, 729-736. Glenn, S., Heyes, A., & Moore, T. (1993) Carbon-Dioxide and Methane Fluxes from Drained Peat Soils, Southern Quebec. Global Biogeochemical Cycles, 7, 247-257. Grace, J. (2004) Understanding and managing the global carbon cycle. Journal of Ecology, 92, 189-202.
52
Chapter 1 Grieve, I.C., Davidson, D.A., & Gordon, J.E. (1995) Nature, Extent and Severity of Soil-Erosion in Upland Scotland. Land Degradation and Rehabilitation, 6, 41-55. Haines-Young, R.H., Barr, C.J., Black, H.I.J., Briggs, D.J., Bunce, R.G.H., Clarke, R.T., Cooper, A., Dawson, F.H., Firbank, L.G., Fuller, R.M., Furse, M.T., Gillespie, M.K., Hill, R., Hornung, M., Howard, D.C., McCann, T., Morecroft, M.D., Petit, S., Sier, A.R.J., Smart, S.M., Smith, G.M., Stott, A.P., Stuart, R.C., & Watkins, J.W. (2000) Accounting for nature: assessing habitats in the UK countryside DETR, London. Hamilton, A. (2000) The characteristics and effects of management fire on blanket bog vegetation in north-west Scotland. PhD Thesis, University of Edinburgh, Edinburgh. Hamilton, A., Legg, C.J., & Zhaohua, L. (1997) Blanket mire research in north-west Scotland: a view from the front. In Blanket mire degradation causes, consequences and challenges (eds J.H. Tallis, R. Meade & P.D. Hulme), pp. 47-53. The Macaulay Land Use Research Institute, Aberdeen, University of Manchester. Hargreaves, K.J., Milne, R., & Cannell, M.G.R. (2003) Carbon balance of afforested peatland in Scotland. Forestry, 76, 299-317. Heathwaite, A.L. (1993) Disappearing Peat Regenerating Peat - the Impact of Climate- Change on British Peatlands. Geographical Journal, 159, 203-208. Hope, D., Billett, M.F., & Cresser, M.S. (1997) Exports of organic carbon in two river systems in NE Scotland. Journal of Hydrology, 193, 61-82. Howard, P.J.A., Loveland, P.J., Bradley, R.I., Dry, F.T., Howard, D.M., & Howard, D.C. (1995) The carbon content of soil and its geographical distribution in Great Britain. Soil Use and Management, 11, 9-15. Hulme, M., Jenkins, G.J., Lu, X., Turnpenny, J.R., Mitchell, T.D., Jones, R.G., Lowe, J., Murphy, J.M., Hassell, D., Boorman, P., McDonald, R., & Hill, S. (2002). Climate Change Scenarios for the United Kingdom: The UKCIP02 Scientific Report.
53
Chapter 1 Tyndall Centre for Climate Change Research, School of Environmental Sciences, University of East Anglia, Norwich, UK. Hulme, P.E. (2005) Adapting to climate change: is there scope for ecological management in the face of a global threat? Journal of Applied Ecology, 42, 784-794. Ingram, H.A.P. (1978) Soil layers in mires: Function and terminology. Journal of Soil Science, 29, 224-227. IPCC (2001) Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change Cambridge University Press, Cambridge. JNCC (1994) Guidelines for the selection of biological SSSIs: bogs. Joint Nature Conservancy Committee, Peterborough. Joabsson, A., Christensen, T.R., & Wallen, B. (1999) Vascular plant controls on methane emissions from northern peat forming wetlands. Trends in Evolution and Ecology, 14, 385-388. Johnson, G.M. & Morris, J.M. (2000a). Scottish Blanket Bog Inventory: Grampian region and adjacent areas - Characterisation of blanket bogs using Landsat Thematic Mapper, Rep. No. Scottish Natural Heritage Research Survey and Monitoring Series. Scottish Natural Heritage, Perth. Johnson, G.M. & Morris, J.M. (2000b). Scottish Blanket Bog Inventory: The Inner Isles and Mid-west coast of Scotland - Characterisation of blanket bogs using Landsat Thematic Mapper, Rep. No. Scottish Natural Heritage Research Survey and Monitoring Series. Scottish Natural Heritage, Perth. Johnson, G.M. & Morris, J.M. (2000c). Scottish Blanket Bog Inventory: The Southeast - Characterisation of blanket bogs using Landsat Thematic Mapper., Rep. No. Research Survey and Monitoring Series. Scottish Natural Heritage, Perth. Johnson, G.M. & Morris, J.M. (2000d). Scottish Blanket Bog Inventory: The Southwest - Characterisation of blanket bogs using Landsat Thematic Mapper, Rep. No.
54
Chapter 1 Scottish Natural Heritage Research Survey and Monitoring Series. Scottish Natural Heritage, Perth. Johnson, G.M. & Morris, J.M. (2001). Scottish Blanket Bog Inventory: Caithness & Sutherland and Orkney - Characterisation of blanket bogs using Landsat Thematic Mapper., Rep. No. Scottish Natural Heritage Research Survey and Monitoring Series. Scottish Natural Heritage, Perth. Johnson, R.H., Tallis, J.H., & Wilson, P. (1990) The Seal-Edge-Coombes, North Derbyshire - a Study of Their Erosional and Depositional History. Journal of Quaternary Science, 5, 83-94. Jowsey, P.C. (1973). Peatlands. In The organic resources of Scotland: their nature and evaluation (ed T. J), pp. 109-121. Oliver and Boyd, Edinburgh. Keeling, C.D. & Whorf, T.P. (2005). Atmospheric CO2 records from sites in the SIO air sampling network. In In Trends: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A. Keppler, F., Hamilton, J.T.G., Braß, M., & Röckman, T. (2006) Methane emission from terrestrial plants under aerobic conditions. Nature, 439, 188-191. King, D. (2005) Climate change: the science and the policy. Journal of Applied Ecology, 42, 779-783. Kuhry, P. (1994) The role of fire in the development of Sphagnum-dominated peatlands in western boreal Canada. Journal of Ecology, 82, 899-910. Legg, C.J., Maltby, E., & Proctor, M.C.F. (1992) The ecology of severe moorland fire on the North York Moors: seed distribution and seedling establishment of Calluna vulgaris. Journal of Ecology, 80, 737-752. Lindsay, R.A. (1995) Bogs: the Ecology Classification and Conservation of Ombrotrophic Mires. Scottish Natural Heritage, Battleby.
55
Chapter 1 Lindsay, R.A., Charman, D.J., Everingham, F., O'Reilly, R.M., Palmer, M.A., Rowell, T.A., & Stroud, D.A. (1988) The Flow Country: The Peatlands of Caithness and Sutherland. Nature Conservancy Council, Peterborough. Lloyd, D., Thomas, K.L., Benstead, J., Davies, K.L., Lloyd, S.H., Arah, J.R.M., & Stephen, K.D. (1998) Methanogenesis and CO2 exchange in an ombrotrophic peat bog. Atmospheric Environment, 32, 3229-3238. Mackay, A.W. & Tallis, J.H. (1996) Summit-type blanket mire erosion in the forest of Bowland, Lancashire, UK: Predisposing factors and implications for conservation. Biological Conservation, 76, 31-44. Maltby, E., Legg, C.J., & Proctor, M.C.F. (1990) The ecology of severe moorland fire on the North York Moors: effects of the 1976 fires, and subsequent surface and vegetation development. Journal of Ecology, 78, 490-518. McHugh, M., Harrod, T., & Morgan, R. (2002) The extent of soil erosion in upland England and Wales. Earth Surface Processes and Landforms, 27, 99-107. Milne, R. & Brown, T.W.A. (1997) Carbon in vegetation and soils of Great Britain. Journal of Environmental Management, 49, 413-433. MISR (1984). Organisation and methods of the 1:250,000 Soil Survey of Scotland. Macaulay Institute for Soil Research, Aberdeen. Monteith, D.T. & Evans, C.D., eds. (2000) UK Acid Waters Monitoring Network: 10 year report. Analysis and interpretation of results April 1988 - March 1998. ENSIS, London. Moore, P.D., Merryfield, D.L., & Price, M.D.R. (1984). The vegetation and development of British mires. In European mires (ed P.D. Moore). Academic Press, London. NCC (1990) Handbook for Phase 1 habitat survey - a technique for environmental audit Nature Conservancy Council, Peterborough.
56
Chapter 1 Oliver, H.R., Bell, B.G., & Clymo, R.S. (1998) The TIGER programme. Atmospheric Environment, 32, 3205-3205. O'Neill, K.P. (2000). Role of bryophyte-dominated ecosystems in the global carbon budget. In Bryophyte Biology (eds A.J. Shaw & B. Goffinet), pp. 344-368. Cambridge University Press, Cambridge. Palmer, S.M., Hope, D., Billett, M.F., Dawson, J.J.C., & Bryant, C.L. (2001) Sources of organic and inorganic carbon in a headwater stream: Evidence from carbon isotope studies. Biogeochemistry, 52, 321-338. Pastor, J., Solin, J., Bridgham, S.D., Updegraff, K., Harth, C., Weishampel, P., & Dewey, B. (2003) Global warming and the export of dissolved organic carbon from boreal peatlands. Oikos, 100, 380-386. Patterson, G. & Russell, A. (2000). Forests and Peatland Habitats. Guideline Note. Forestry Commission, Edinburgh. Peterken, G.F. (1996) Natural woodland ecology and conservation in northern temperate regions Cambridge University Press, Cambridge. Quarmby, N.A., Johnson, G., & Morris, J.M. (1999). Scottish Blanket Bog Inventory: The Shetland Islands - Characterisation of blanket bogs using Landsat Thematic Mapper, Rep. No. Scottish Natural Heritage Commissioned Report (unpublished report). Scottish Natural Heritage, Perth. Rawes, M. (1981) Further results of excluding sheep from high-level grasslands in the north Pennines. Journal of Ecology, 69, 651-669. Rawes, M. (1983) Changes in two high altitude blanket bogs after the cessation of sheep grazing. Journal of Ecology, 71, 219-235. Rawes, M. & Hobbs, R. (1979) Management of semi-natural blanket bog in the northern Pennines. Journal of Ecology, 67, 789-807.
57
Chapter 1 Robertson, R.A. (1971) Nature and extent of Scottish peatlands. Acta Agralia Fennica, 123. Robertson, R.A. & Jowsey, P.C. (1968) Peat resources and development in the United Kingdom. In Third International Peat Congress, pp. 13-14, Quebec. Scott, M.J., Jones, M.N., Woof, C., Simon, B., & Tipping, E. (2001) The molecular properties of humic substances isolated from a UK upland peat system - A temporal investigation. Environment International, 27, 449-462. Scott, M.J., Jones, M.N., Woof, C., & Tipping, E. (1998) Concentrations and fluxes of dissolved organic carbon in drainage water from an upland peat system. Environment International, 24, 537-546. Shaw, S.C., Wheeler, B.D., Kirby, P., Phillipson, P., & Edinutids, R. (1996). Literature review of the historical effects of burning and grazing of blanket bog and upland wet heath, Rep. No. English Nature Research Reports No. I72. English Nature, Peterborough. St Louis, V.L., Partridge, A.D., Kelly, C.A., & Rudd, J.W.M. (2003) Mineralization rates of peat from eroding peat islands in reservoirs. Biogeochemistry, 64, 97-110. Steiner, G. (1997). The bogs of Europe. In Conserving Peatlands (eds L. Parkyn, R.E. Stoneman & H.A.P. Ingram), pp. 25-34. CAB International, Wallingford. Stevenson, A.C., Jones, V.J., & Battarbee, R.W. (1990) The Cause of Peat Erosion a Paleolimnological Approach. New Phytologist, 114, 727-735. Stewart, A.J.A. & Lance, A.N. (1991) The effects of moor-draining on the hydrology and vegetation of northern Pennine blanket bog. Journal of Applied Ecology, 28, 1105-1117. Stewart, J.M., Birnie, A.C., & Mitchell, B.D. (1966) Characterization of a Peat Profile by Thermal Methods. Agrochimica, 11, 92-&.
58
Chapter 1 Tallis, J.H. (1959) Studies in the Biology and Ecology of RhacomitriumLanuginosum Brid .2. Growth, Reproduction and Physiology. Journal of Ecology, 47, 325-350. Tallis, J.H. (1964) Studies on Southern Pennine Peats .2. The Pattern of Erosion. Journal of Ecology, 52, 333-344. Tallis, J.H. (1965) Studies on Southern Pennine Peats .4. Evidence of Recent Erosion. Journal of Ecology, 53, 509-520. Tallis, J.H. (1973) Studies on Southern Pennine Peats .5. Direct Observations on Peat Erosion and Peat Hydrology at Featherbed Moss, Derbyshire. Journal of Ecology, 61, 1-22. Tallis, J.H. (1985) Mass Movement and Erosion of a Southern Pennine Blanket Peat. Journal of Ecology, 73, 283-315. Tallis, J.H. (1987) Fire and Flood at Holme Moss - Erosion Processes in an Upland Blanket Mire. Journal of Ecology, 75, 1099-1129. Tallis, J.H. (1994) Pool and hummock patterning in a southern Peninne blanket mire. II. The formation and erosion of the pool system. Journal of Ecology, 82, 789-804. Tallis, J.H. (1995) Climate and erosion signals in British blanket peats: The significance of Rhacomitrium lanuginosum remains. Journal of Ecology, 83, 10211030. Tallis, J.H. (1997a) The pollen record of Empetrum nigrum in southern Pennine peats: implications for erosion and climate change. Journal of Ecology, 85, 455-465. Tallis, J.H. (1997b) The Southern Pennine experience: an overview of blanket mire degradation. In Blanket Mire Degradation, causes, consequences and challenges. MacAlauy Land Use Research Institute, University of Manchester.
59
Chapter 1 Tallis, J.H. & Livett, A. (1994) Pool and hummock patterning in a southern Peninne blanket mire. I. Stratigraphic profiles for the last 2800 years. Journal of Ecology, 82, 775-788. Tipping, E., Woof, C., Rigg, E., Harrison, A.F., Ineson, P., Taylor, K., Benham, D., Poskitt, J., Rowland, A.P., Bol, R., & Harkness, D.D. (1999) Climatic influences on the leaching of dissolved organic matter from upland UK Moorland soils, investigated by a field manipulation experiment. Environment International, 25, 8395. Tucker, G. (2003). Review of the impacts of heather and grassland burning in the uplands on soils, hydrology and biodiversity, Rep. No. English Nature Research Reports No. 550. English Nature, Peterborough. Vitt, D.H. (2000). Peatlands: ecosystems dominated by bryophytes. In Bryophyte Biology (eds A.J. Shaw & B. Goffinet), pp. 312-343. Cambridge University Press, Cambridge. von Post, L. & Sernander, R. (1910) Pflanzen-physiogno-mische Studien auf Torfmooren in Nirken. In XI International Geological congress: Excursion Guide No. 14 (A7), pp. 1-48, Stockholm. Waddington, J.M. & McNeill, P. (2002) Peat oxidation in an abandoned cutover peatland. Canadian Journal of Soil Science, 82, 279-286. Waddington, J.M. & Roulet, N.T. (2000) Carbon balance of a boreal patterned peatland. Global Change Biology, 6. Warburton, J. (2003) Wind-splash erosion of bare peat on UK upland moorlands. Catena, 52, 191-207. Warburton, J., Higgit, D., & Mills, A. (2003) Anatomy of a Pennine peat slide, northern England. Earth Surface Processes and Landforms, 28, 457-473. Warburton, J., Holden, J., & Mills, A.J. (2004) Hydrological controls of surficial mass movements in peat. Earth-Science Reviews, 67, 139-156.
60
Chapter 1 Welch, D. & Rawes, M. (1965) The herbage production of some Pennine grasslands. Oikos, 16, 39-47. Whalen, S.C. & Reeburgh, W.S. (1992) Interannual variations in tundra methane emission a 4-year time series at fixed sites. Global Biogeochemical Cycles, 6, 139159. Wheeler, B.D. & Shaw, S.C. (1995) Restoration of damaged peatlands with particular reference to lowland raised bogs affected by peat extraction HMSO, London. Whiting, G.J. & Chanton, J.P. (1993) Primary production control of methane emission from wetlands. Nature, 364, 794-795. Wishart, D. & Warburton, J. (2001) An assessment of blanket mire degradation and peatland gully development in the Cheviot Hills, Northumberland. Scottish Geographical Journal, 117, 185-206. Worrall, F. & Burt, T. (2005) Predicting the future DOC flux from upland peat catchments. Journal of Hydrology, 300, 126-139. Worrall, F., Burt, T., & Adamson, J. (2004a) Can climate change explain increases in DOC flux from upland peat catchments? Science of the Total Environment, 326, 95112. Worrall, F., Burt, T., & Shedden, R. (2003a) Long term records of riverine dissolved organic matter. Biogeochemistry, 64, 165-178. Worrall, F., Harriman, R., Evans, C.D., Watts, C.D., Adamson, J., Neal, C., Tipping, E., Burt, T., Grieve, I., Monteith, D., Naden, P.S., Nisbet, T., Reynolds, B., & Stevens, P. (2004b) Trends in dissolved organic carbon in UK rivers and lakes. Biogeochemistry, 70, 369-402. Worrall, F., Reed, M., Warburton, J., & Burt, T. (2003b) Carbon budget for a British upland peat catchment. The Science of the Total Environment, 312, 133-146.
61
Chapter 1 Yeo, M. (1998) Blanket Mire Degradation in Wales. In Blanket mire degradation causes, consequences and challenges (eds J.H. Tallis, R. Meade & P.D. Hulme). The Macaulay Land Use Research Institute, Aberdeen, University of Manchester. Younger, P.L. & McHugh, M. (1995) Peat Development, Sand Cones and Paleohydrogeology of a Spring- Fed Mire in East Yorkshire, UK. Holocene, 5, 5967.
62
Chapter 2 Chapter 2: Peatland gaseous carbon fluxes and land management: searching for a paradigm. 2.1 Introduction Carbon flux research is important for the parameterisation of climate change models, understanding ecosystem response to climate change and informing government policy (Grace, 2004). This type of research frequently applies a bottom up approach where smaller scale research is scaled up to the landscape or higher scales (Grace et al., 2001). Meta-analysis has long been used in clinical and social science studies especially when informing wider society e.g. (Roberts et al., 2002; Altun & Arici, 2006) and is gaining in popularity in ecology (Osenberg et al., 1999; Gates, 2002) where it has recently been incorporated into a number of reviews of carbon dynamics (Peterson et al., 1999; Johnson & Curtis, 2001; Guo & Gifford, 2002; Wang & Curtis, 2002; Long et al., 2004; van Kooten et al., 2004; Manley et al., 2005; Ogle et al., 2005). Peatland ecosystems in the boreal region store large amounts of carbon (Clymo et al., 1998) and the interactions between these ecosystems and the atmosphere are important to climate change research. The most significant greenhouse gases in terms of ombrotrophic peatlands are CO2 and CH4. On the other hand, N2O appears less significant but may be more prevalent in more minerotrophic peatlands (Byrne et al., 2004). In the past few decades technological and analytical advances such as eddy co-variance have allowed the estimation of gaseous fluxes of CO2 and CH4 from ecosystems at fine temporal and large spatial scales (Beverland et al., 1996; Beswick et al., 1998). These have allowed informed estimates of the greenhouse dynamics of northern peatlands to be made. Blanket bog is the most important peatland habitat and the most extensive seminatural land habitat in the UK (Lindsay, 1995). The UK holds 10-15% of the total world area of this habitat (Lindsay, 1995) but is only approximately 0.16% of the global land mass, emphasising the importance of peatlands in the UK. The development of blanket bog is a function of past and present environmental factors (e.g. climate, geology, geomorphology) and of the nature, intensity and history of human impact (Steiner, 1997).
Threats to these peatland ecosystems include
63
Chapter 2 drainage, agricultural improvement, burning, the effects of large herbivores, peat extraction and climate change. Peatland research in the UK has at least a century long history. However, UK peatlands have been subjected to several centuries of land management practices such as burning and grazing (Shaw et al., 1996). Further, the UK’s climate is oceanic and therefore climate change and ecosystem responses to climate change are likely to be different from those of the north American and European continents. Therefore, parameterisation of UK climate change models, understanding peatland response to climate change and informing government policy is likely to require a UK perspective. In this scenario a meta-analytical methodology to the analysis of peatland carbon fluxes and management would seem an ideal approach. Here I attempt to apply a semi-quantitative approach to review gaseous CO2 and CH4 fluxes from UK peatlands in order to: 1. summarise previous work, 2. provide evidence of how management influences carbon fluxes in UK peatlands, and 3.
indicate areas of study where research may be lacking.
2.2 Methods Published literature on the effects of management on the gaseous carbon fluxes of blanket bog was searched using the following online bibliographic databases available through the University of Edinburgh Library: ISI Web of Knowledge JSTOR INGENTA ZETOC Index to theses in Great Britain and Ireland As well as the above databases keywords were also used as parameters for searches using the internet search engines Google, Google Scholar and Scirus. A list of keywords used (in various combinations) as search parameters for these databases are shown below in Table 2.1.
64
Chapter 2 Table 2.1: Examples of keywords used in literature review searches. blanket bog
DOC
moor and moss
bog
Erica
muir
burn
Eriophorum
muirburn
burning
erosion
peat
Calluna
fire
peatland
carbon
grazing
restoration
carbon dioxide
heath
Scotland
cattle
heather
sheep
CH4
hummocks hollows etc.
Sphagnum
CO2
mire
UK
deer
Molinia
upland
dissolved organic carbon
moor
wetland
In addition to the keywords in Table 2.1 certain authors were used in more specific searches, e.g., Clymo. The reference lists within journal papers were also investigated to identify any relevant papers. Also, of particular value for the older literature was ‘Peatland Ecology in the British Isles: a Bibliography’ (Field, 1981). However, it may be possible that certain references have been overlooked the main gaps are likely to be unpublished studies or reports. There is a large array of molar and mass units reported in literature but authors do not always explicitly state which substance units pertain to, CO2 or CO2-C and CH4 or CH4-C. Unless authors have stated units, the approach adopted here is to make the assumption that when examining fluxes of CO2 units are defined in terms of fluxes of CO2 and when examining CH4 they are defined in terms of CH4. As many data points as possible were included from each study and all are given in the tables in the appendix.
65
Chapter 2 2.3 Results Table 2.2: Number and characteristics of gaseous CO2 and CH4 flux studies conducted in the UK from a review of papers (Clymo & Reddaway, 1971, 1972; Choularton et al., 1995; Clymo & Pearce, 1995; Fowler et al., 1995a; Fowler et al., 1995b; Nedwell & Watson, 1995; Beverland et al., 1996; Chapman & Thurlow, 1996; Fowler et al., 1996; Gallagher et al., 1996; Beswick et al., 1998; Chapman & Thurlow, 1998; Daulaut & Clymo, 1998; Hargreaves & Fowler, 1998; Lloyd et al., 1998; MacDonald et al., 1998; Moncrieff et al., 1998; Hughes et al., 1999; Freeman et al., 2002; Gauci et al., 2002; Hargreaves et al., 2003; Beckmann et al., 2004) using a keyword searches of bibliographic databases. * Note: does not necessarily sum to total number of studies because some papers used multiple methods. N/S - not stated. Gas
No. Studies
Country Method*
No.
Bog Type
Management Winter
Sites CO2 8 3 respiration
7 Scot.
3 peat cores/lab
1 Eng.
1 conditional
only
6
included 4 blanket
8 N/S
2 raised
3 included 4 not
sampling
included
3 static chamber
2 not stated
2 eddy covariance CH4 19
17 Scot.
6 peat core/lab
1 Eng.
2 conditional
1 soligenous
8 not
1 Wales
sampling
gully mire
included
6 static chamber
9 blanket
6 not stated
11
1 raised
19 N/S
5 included
6 eddy covariance 4 aircraft 3 vertical profile 2 tethered balloon 2 nocturnal box 2 flux gradient
Table 2.2 summarises the work found by this review. A total of eight CO2 studies were found but of these three were respiration only studies. There is a bias in terms of the countries studied towards Scotland with only that of Clymo and Reddaway (1971 & 1972) from England. The methodology employed is fairly evenly split between static chambers, peat cores and eddy covariance/conditional sampling. These methods have employed a variety of scales from < 1m2 (chambers, cores) to > 1 km2 (eddy covariance). Although there are 8 studies, only 6 sites have been
66
Chapter 2 sampled, therefore, some sites have been re-sampled and not always by the same authors. There are double the number of blanket bog sites (4) compared to raised bog (2) sampled. None of the studies stated the type of management of the bog and winter only appears to have been sampled in half of the studies. A total of nineteen studies were found by this review to have examined fluxes of CH4. Seventeen were in Scotland, one in England and one in Wales again reflecting country bias. There seem to be a larger array of methods employed and a wide variety of scales from < 1m2 to almost the entire north of Scotland (aircraft) (Fowler et al., 1996; Gallagher et al., 1996; Beswick et al., 1998). The numbers of sites used are again less than the number of studies indicating re-use of sites for subsequent studies. A much higher of proportion of blanket bog is represented with nine sites, with one raised bog and a soligenous gully mire also sampled. As with the CO2 studies there is no information on site management and winter is also under represented with only five of the nineteen studies covering this season. Table 2.3 and Figure 2.1 show mean CO2 fluxes, standard error, mean net flux (light and dark) and sites sampled for each of the studies examined by this review. Dark fluxes range from 0.06 to 1.389 µmol CO2 m-2 s-1 and light fluxes from -5.556 to 0.704 µmol CO2 m-2 s-1. The reported values give an overall mean net flux of -0.640 (se 0.925) µmol CO2 m-2 s-1 appearing to indicate an overall sink for CO2. Figure 2.1 also shows outlying points in grey all of which come from the study of Beverland et al, (1996). Figure 2.1 and Table 2.3 also indicate the paucity of studies reporting CO2 fluxes in the light, numbering only three.
67
Chapter 2 Table 2.3 Mean carbon dioxide flux results from published papers examined by this report; units are µmol CO2 m-2 s-1. Note; n in column 7 relates to the number of reported values from which a study mean was derived. * Value not reported. Study Reference
Site
Light Mean CO2
No. 1
n Mean Net
Dark
flux
Dark
0.097
0.037
3
Light
-0.014
0.068
3
Dark
1.389
1.389
2
Light
-5.556
2.778
2
Glensaugh
Dark
0.195
0.037
2
N/A
Clymo and
Ellergower
Dark
0.060
0.026
2
N/A
Pearce (1995)
Moss
Clymo and
Moor
House Dark
0.119
0.018
3
N/A
Reddaway
Burnt Hill Dark
0.611
*
1
-0.389
Light
-1.00
*
1
*
*
1
-0.002 1.276
Beckman et al Ellergower (2004)
2
SE
flux 0.082
Moss
Beverland et al Loch More
-4.167
(1996)
3
Chapman and Thurlow (1996)
4 5
(1971 and 1972) 6
Fowler et al
Loch More
(1995a)
7 8
Hargreaves et al Auchencorth Net (2003)
Moss
rate
Lloyd et al
Ellergower
Dark
0.572
0.453
2
(1998)
Moss Light
0.704
0.557
2
0.925
5
Mean
Scottish sites Net rate
68
-0.640
carbn dioxide flux rate
Chapter 2
Dark Light Net rate 0
-4
-8
1
2
3
4
5
6
7
8
Study No.
Figure 2.1: Carbon dioxide flux (µmol CO2 m-2 s-1) against study number from Table 11, outlying data points are shown in grey. As there are more CH4 studies Table 2.4 and Figure 2.2 summarise the mean CH4 fluxes (se) by site rather than by each paper examined. What is immediately apparent from Table 2.4 is that some sites are more frequently reported than others. Loch More has eight published results, four from Ellergower Moss, three from Caithness and Strathy Bog and the rest of the sites are reported once. Values range from 0.01 µmol CH4 m-2 s-1 at Moor House in north England to 0.131 µmol CH4 m-2 s-1 at Cerrig-yr-Wyn in Wales. Overall mean CH4 flux 0.029 (se 0.01) µmol CH4 m-2 s-1. CH4 fluxes appear to be less prone to outliers except the values reported from Cerrigyr-Wyn in Wales, which are high in comparison to the rest.
69
Chapter 2 Table 2.4: Site mean methane flux results from published papers examined by this report; units are µmol CH4 m-2 s-1. Note; n in column 4 relates to the number of reported values from a site from which the mean is derived. (Clymo & Reddaway, 1971, 1972; Choularton et al., 1995; Clymo & Pearce, 1995; Fowler et al., 1995a; Fowler et al., 1995b; Nedwell & Watson, 1995; Beverland et al., 1996; Chapman & Thurlow, 1996; Fowler et al., 1996; Gallagher et al., 1996; Beswick et al., 1998; Chapman & Thurlow, 1998; Daulaut & Clymo, 1998; Hargreaves & Fowler, 1998; Lloyd et al., 1998; MacDonald et al., 1998; Moncrieff et al., 1998; Hughes et al., 1999; Freeman et al., 2002; Gauci et al., 2002; Hargreaves et al., 2003; Beckmann et al., 2004) CH4 site
Mean flux SE
n
Bad a Cheo
0.024
*
1
Caithness
0.034
0.009
3
Cerrig-yr-Wyn
0.131
*
1
Ellergower Moss
0.016
0.009
4
Loch Calium
0.014
*
1
Loch More
0.013
0.002
8
Moidach More
0.020
*
1
Moor House
0.010
*
1
North Scotland
0.013
*
1
Potree to Wick
0.014
*
1
Strathy Bog
0.032
0.023
3
Mean of all sites
0.029
0.010
11
70
methane flux rate
Chapter 2
0.12
0.07
Strathy Bog
Potree to Wick
North Scotland
Moor House
Moidach More
Loch More
Loch Calium
Ellergower Moss
Cerrig-yr-Wyn
Caithness
Bad a Cheo
0.02
Figure 2.2: Site mean methane flux results from published papers examined by this report, units of flux are µmol CH4 m-2 s-1, error bars represent standard error. Site mean methane flux results from published papers examined by this report; units are µmol CH4 m-2 s-1. Overall mean is 0.029 µmol CH4 m-2 s-1 (0.01). Figure 2.3 examines some of the data from the UK peatlands where both CO2 and CH4 data are available, implemented in the model of Whiting and Chanton (2001). This model presents the molar ratio of CH4/CO2 against the molar global warming potential of methane (GWPM) over time. The greenhouse compensation point represents a line whereby, the emission of CH4 is balanced by the molar uptake of CO2, and therefore, any data lying along this line is greenhouse neutral.
71
Chapter 2
Loch More mean
Loch More 1
25
All Data and Loch More 2
Beverland
Source
20 years
Greenhouse compensation point
GWPM (mol/mol)
20
Sink
15
10
100 years 5
500 years 0 -0.01
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
(CH4 / CO2) emission exchange ratio (mol/mol)
Figure 2.3: Model relating CH4/CO2 emission ratio to Global Warming Potential (GWPM) and time for UK peatlands. Loch More 1 is calculated from the high and low values reported by Beverland et al., (1996); Loch More 2 is calculated from Fowler et al., (1995a), the Beverland ratio is calculated from reported annual sink source data (Beverland et al., 1996), Loch More mean is the mean of all Loch More data, and All data represents the ratio calculated from mean all available values found in this review. Figure 2.3 suggests that from the available data, when both CH4 and CO2 are taken account of, UK peats appear to be sinks for carbon in terms of global warming potential. Only the All data and Loch More 2 are marginal sinks over the 20-year scenario. However, due to the limitations of the carbon dioxide data found by this review, the results presented in Figure 2.3 can only be regarded as illustrative. 2.4 Discussion 2.4.1 Fluxes of CO2 The evidence given above would appear to indicate that peatlands in the UK may be a sink for atmospheric CO2 and the overall mean figure of -0.640 µmol CO2 m-2 s-1 would seem to offer support for this. However, there are very few studies, only 8 in
72
Chapter 2 total, only 5 of these recorded light and dark fluxes and only 1 of these includes winter. Further, this mean is influenced by some extreme values highlighted by Figure 2.1. The extreme values reported by Beverland et al., (1996) arise because the authors reported high and low values, and the study was conducted in the height of summer when rates of exchange are at their greatest. Removing these and then recalculating the mean is unsatisfactory because the mean would then be dominated by the Ellergower Moss results of Lloyd et al., (1998) and Beckman et al., (2004). This is unsatisfactory because Ellergower is a raised bog not a blanket bog and therefore not representative of the UK peatland habitat as a whole, and these studies both reported CO2 emissions in illuminated laboratory controlled conditions, intuitively this would appear to be unrepresentative. This would leave the reported flux of Hargreaves et al., (2003) as the only representative measure for blanket peat CO2 flux rates. This, though, is a partly modelled value using climate data from Newton Stewart to derive a net flux rate for Auchencorth Moss approximately 130 km to the south-west not actual climate data from the site. There appears then to be no satisfactory mean value for the gaseous flux of CO2 from UK peatlands. 2.4.2 Fluxes of CH4 It is apparent that there is more published information on CH4 fluxes from UK peatlands than fluxes of CO2. From the total of nineteen studies from eleven different sites all reporting emissions of methane, an overall mean emission is 0.029 µmol (se 0.01) CH4 m-2 s-1. Only the values reported from Cerrig-yr-Wyn in Wales appear to be unusually large (Figure 3) but this may be due to the influence of groundwater in this slightly different habitat (soligenous mire). However, only six of the nineteen explicitly state that winter was included and this would appear to be an under represented season. 2.4.3 UK Peatlands, overall C source or sink? Beverland et al., (1996) conclude from their results that the site would represent an annual sink of –0.5 Mt C for UK peatlands. Given the limitations of their study and the very high error variance, this is unlikely to be a reliable estimate. Hargreaves et al., (2003) give a net rate of -0.25 t C ha-1 yr-1 but this is also unlikely to be reliable
73
Chapter 2 because of the use of remote climate data in modelling. We must also state that the situation illustrated by Figure 2.3, and the mean values of -0.640 CO2 m-2 s-1 and 0.029 µmol CH4 m-2 s-1 obtained from this review, are unlikely to be reliable estimates due to the paucity of results, susceptibility to extreme values and the seasonal limitations of current research. 2.4.4 Representation of sampled sites The country bias found in that more sites are situated in Scotland than England and Wales is expected since this is where the majority of peatlands are found in the UK (Lindsay, 1995). However, it is necessary to ask whether the sites where fluxes have been reported are representative of the entire blanket bog situation in the UK. Table 2.3 indicates that nine studies have sampled a total of 6 sites for CO2 and Table 2.4 nineteen studies from eleven sites or areas for CH4. Given that the blanket bog covers 1.9 million ha it is unlikely that these sites are adequate. Also two sites are raised rather than a blanket bog (Clymo & Pearce, 1995; Nedwell & Watson, 1995; Lloyd et al., 1998; Gauci et al., 2002; Beckmann et al., 2004). Although the vegetation of raised and blanket bog has similarities, the hydrologies are different and the accumulation of peat (hence carbon fixation) has been much greater historically in most raised than in blanket bogs. Even when entire geographical areas are reported using aircraft, the duration of these studies is extremely short, 1 day, with a total of 3 different days sampled in different seasons and years; 24/7/92, 3/6/93, and 29/11/94 (Fowler et al., 1996; Gallagher et al., 1996; Beswick et al., 1998). Also, the assumption that the sampling technique has adequately represented the natural variation present within the site is unlikely to have been met. Eddy co-variance is claimed to report average emissions representative of areas of km2, however, it should be remembered that the sample size in eddy-covariance studies is usually 1 tower, in other words there is no replication; there is therefore a reliance on technology to deliver accurate results with no estimation of spatial variation or precision. Chamber or peat core studies on the other hand usually have much higher replication but cover areas of usually less than 1 m2. Monolith and peat core studies are further complicated by disturbance and the fact that they are usually conducted in the laboratory, i.e. not in the climatic condition in which they were found. Therefore,
74
Chapter 2 present gaseous carbon flux research on peatlands in the UK cannot be regarded as representative at local, regional or national levels. 2.4.5 Management As one of the primary objectives of this review was to examine carbon fluxes in relation to management it is disappointing to report that none of the research examined during this review stated site management. It is therefore not possible to apply a full meta-analysis investigation into the effects of management on gaseous carbon flux at present. However, there has been some recent carbon accumulation peat core work on a long-term burning and grazing experiment at Hard Hill, on the Moor House NNR in the Pennines in England (Garnett, 1998; Garnett et al., 2000). The Hard Hill experiment is a split plot design with three burning treatments not burnt since 1954, a 10-year burn rotation and a 20-year burn rotation. These are then split between grazed and ungrazed plots. This experiment has been running since the 1954 and although Garnett et al., (2000) did not examine all treatments, they conclude from the core data that burning on the 10 year rotation has an adverse effect on carbon accumulation, but there was no detectable effect of grazing probably due to the low stock rates at Moor House. However, as peat cores integrate peat accumulation over longer periods it is difficult to compare this type of data in terms of gaseous flux data. The reasons for the lack of management details may firstly be because the primary goals of the studies were not to examine management. However, given that all the peatlands in the UK are managed to varying extent (see Chapter 1) it would seem amiss not to include even a cursory description of site management. This would be more difficult for the larger scale aircraft studies but should not be a problem for the smaller scale methods of chambers, peat cores and eddy covariance. It is only possible to speculate on further reasons for this omission but there may also be a misguided view that there are peatlands in the UK that are not managed and can be described as ‘pristine’. Hargreaves et al., (2003) clearly describe Auchencorth Moss in the Scottish Borders as an undisturbed peatland; highly unlikely for a site a few miles from the capital of Scotland and nestled in an area long populated and exploited for agriculture Gauci et al., (2002) describe the raised bog sampled as
75
Chapter 2 ‘pristine’ again this is highly unlikely given that raised bogs are some of the most exploited peatland habitats in the UK. Although gradations in habitat condition undoubtedly exist, the existence of unmanaged peatlands in the UK is questionable. Nevertheless, future carbon flux research in the UK should include descriptions of management even if this is not the primary goal of the research not only to allow adequate evaluation of results but also to allow future reviews to compile results in terms of management. 2.4.6 Climate change: models, ecosystem response and government policy As the information available on gaseous carbon flux data in the UK is sparse, it would seem prudent to ask what options are available if data are to be incorporated into climate change models, or for informing ecosystem response, or even informing government policy. Given the evidence presented in this review it would seem the options are limited to either extrapolation beyond the bounds of the studies or collation of fluxes from other areas such as Canada or Scandinavia. Both extrapolation and collation of fluxes from other areas are undesirable for the following reasons. Extrapolation to arrive at estimates for fluxes of CO2 and or CH4 from present data for UK peatlands requires the acceptance of unrealistic assumptions. As detailed above currently spatial and temporal variation are all inadequately represented. This extrapolative approach then would require further research. This may be compounded by the insistence of some funding bodies and some editorial policy that requires research to be novel, this is at odds with attaining the goal of adequate representation, since it leads to the proliferation of quasi-replicated studies and experiments (Palmer, 2000) instead of the required ‘true’ replication through space and time. The use of data from others areas would seem the only sensible option at present but is also undesirable because firstly as stated above the UK has an oceanic climate unlike the more continental climate of other areas. Further, permafrost studies are not applicable in the UK as the UK does not have any permafrost peatlands and the responses of these systems are likely to differ because predicted temperature rises are believed to be more extreme in more northerly latitudes (IPCC, 1996). Therefore the
76
Chapter 2 amplitude of permafrost boundary variation is likely to have more profound consequences on CO2 and CH4 emissions than those from UK peatlands from predicted climate scenarios. Most importantly, unlike continental peatlands and other northern boreal peatlands UK peatlands have been subjected to deliberate management practices for many centuries and consequently UK peatland ecosystems are in no way pristine or undisturbed. The peat in the UK may therefore differ not only biologically but also physically from those on the continent because of the history of these management practices. This may have important consequences. It is therefore important that future carbon flux research in the UK addresses management issues. 2.4.7 Peatland carbon flux research: a global context The UK may only be approximately 0.16% of the terrestrial biosphere but 10-15% of the total world area of blanket bog is located in the UK (Lindsay, 1995). Historically the UK has made important contributions to gaseous flux research. Indeed Clymo and Reddaway (1971 & 1972) made what may have been the first ever attempt at quantifying CH4 fluxes at Moor House. The TIGER programme provided continuity through to the late 1990's on peatland research and gaseous carbon fluxes in the UK (Oliver et al., 1998). This initial impetus appears to have lapsed in the UK at least for peatland ecosystems, although the Scottish Executive are funding an organic soils modelling project. In other areas such as the north American and European continents peatland gaseous flux research has continued and have helped to elucidate the relationships between environmental controls, the impacts of forestry, drainage and restoration on gas fluxes in peatlands (Billings et al., 1982; Crill et al., 1992; Dise, 1992; Martikainen et al., 1992; Oechel et al., 1993; Whiting & Chanton, 1993; Bubier, 1995; Christensen et al., 1996; Waddington et al., 1996; Bridgham et al., 1999; Christensen et al., 1999; Joabsson et al., 1999; Komulainen et al., 1999; Tuittila, 2000; Aurela et al., 2001; Aurela et al., 2002; Blodau, 2002). The importance of CH4 fluxes from peatlands to the global carbon budget is well evidenced (Gorham, 1991). There are strong links between water table and vegetation on CH4 fluxes, CH4 is oxidised in the acrotelm and research examining the links between water table and vegetation have shown some peatland types to be sinks
77
Chapter 2 and others sources (Bubier et al., 1995; Clymo & Pearce, 1995). Although techniques for measuring continuous CO2 have been used for a while, techniques for the continuous measurement of CH4 are only just becoming cost effective and more widely available. Previously campaign measurements were possible (Beverland et al., 1996). Now tunable diode lasers (TDL) are available that make fast automatic measurements, so that CH4 can be measured by eddy covariance. This kind of research will be vital to complement the CO2 eddy covariance work and elucidate management relationships in peatlands. Further CH4 research is also required to help clarify recent controversy showing that plants even when aerobic, emit methane (Keppler et al., 2006). However the findings remain controversial and are lacking in a biological explanation. What is clear is that there is a continuing and fast developing research base in which the UK appears to be at present lagging behind. 2.5 Conclusions Current research does not allow adequate estimation of gaseous carbon fluxes from peatland ecosystems in the UK. Also the influence of management of gaseous carbon fluxes is lacking. There is an urgent need for further research not only to address this but also to address the lack of spatial and temporal evidence. This has implications for UK climate change models, UK peatland ecosystem response to climate change and UK government policy. Finally research opportunities exist for the elucidation of disturbance effects on peatland gaseous fluxes on large scales that have implications on global carbon dynamics due to emerging technology. References Altun, B. & Arici, M. (2006) Salt and blood pressure: Time to challenge. Cardiology, 105, 9-16. Aurela, M., Laurila, T., & Tuovinen, J.P. (2002) Annual CO2 balance of a subarctic fen in northern Europe: Importance of the wintertime efflux. Journal of Geophysical Research-Atmospheres, 107, art. no.-4607. Aurela, M., Laurila, T., & Tuovinen, J.-P. (2001) Seasonal CO2 balances of a subarctic mire. Journal of Geophysical Research, 106, 1623-1637.
78
Chapter 2 Beckmann, M., Sheppard, S.K., & Lloyd, D. (2004) Mass spectrometric monitoring of gas dynamics in peat monoliths: effects of temperature and diurnal cycles on emissions. Atmospheric Environment, 38, 6907-6913. Beswick, K.M., Simpson, T.W., Fowler, D., Choularton, T.W., Gallagher, M.W., Hargreaves, K.J., Sutton, M.A., & Kaye, A. (1998) Methane emissions on large scales. Atmospheric Environment, 32, 3283-3291. Beverland, I.J., Moncrieff, J.B., O'Neill, D.H., Hargreaves, K.J., & Milne, R. (1996) Measurement of methane and carbon dioxide fluxes from peatland ecosystems by the conditional sampling technique. Quarterly Journal of the Royal Meteorological Society, 122, 819-838. Billings, W.D., Luken, J.O., Mortensen, D.A., & Peterson, K.M. (1982) Arctic tundra: a source or sink for atmospheric carbon dioxide in a changing environment? Oecologia, 53, 7-11. Blodau, C. (2002) Carbon cycling in peatlands - A review of processes and controls. Environmental Reviews, 10, 111-134. Bridgham, S.D., Pastor, J., Updegraff, K., Malterer, T.J., Johnson, K., Harth, C., & Chen, J. (1999) Ecosystem control over temperature and energy flux in northern peatlands. Ecological Applications, 9, 1345-1358. Bubier, J.L. (1995) The relationship of vegetation to methane emission and hydrochemical gradients in northern peatlands. Journal of Ecology, 83, 403-420. Bubier, J.L., Moore, T.R., & Juggins, S. (1995) Predicting methane emissions from bryophyte distribution in northern Canadian peatlands. Ecology, 76, 677-693. Byrne, K.A., Chojnicki, B., Christensen, T.R., Drösler, M., Freibauer, A., Friborg, T., Frolking, S., Lindroth, A., Mailhammer, J., Malmer, N., Selin, P., Turunen, J., Valentini, R., & Zetterberg, L. (2004). EU Peatlands: Current Carbon Stocks and Trace Gas Fluxes. Carbo Europe GHG, Lund.
79
Chapter 2 Chapman, S.J. & Thurlow, M. (1996) The influence of climate on CO2 and CH4 emissions from organic soils. Agricultural and Forest Meteorology, 79, 205-217. Chapman, S.J. & Thurlow, M. (1998) Peat respiration at low temperatures. Soil Biology & Biochemistry, 30, 1013-1021. Choularton, T.W., Gallagher, M.W., Bower, K.N., Fowler, D., Zahniser, M., & Kaye, A. (1995) Trace gas flux measurements at the landscape scale using boundarylayer budgets. Philosophical Transactions of the Royal Society of London, A, 357369. Christensen, T.R., Jonasson, S., Callaghan, T.V., Havstrom, M., & Livens, F.R. (1999) Carbon cycling and methane exchange in Eurasian tundra ecosystems. Ambio, 28, 239-244. Christensen, T.R., Prentice, I.C., Kaplan, J., Haxeltine, A., & Stich, S. (1996) Methane flux from northern wetlands and tundra. Tellus, 48B, 652-661. Clymo, R.S. & Pearce, D.M.E. (1995) Methane and Carbon-Dioxide Production in, Transport through, and Efflux from a Peatland. Philosophical Transactions of the Royal Society of London Series a-Mathematical Physical and Engineering Sciences, 351, 249-259. Clymo, R.S. & Reddaway, J.F. (1971) Productivity of Sphagnum (bog moss) and peat accumulation. Hidrobiologia, 12, 181-192. Clymo, R.S. & Reddaway, J.F. (1972). A tentative dry matter balance for the wet blanket bog on Burnt Hill Moor House NNR, Rep. No. Aspects of the Ecology of the Northern Pennines. Occasional Papers No. 3. Nature Conservancy. Clymo, R.S., Turunen, J., & Tolonen, K. (1998) Carbon accumulation in peatland. Oikos, 81, 368-388. Crill, P., Bartlett, K., & Roulet, N.T. (1992) Methane flux from boreal peatlands. Suo, 43, 173-182.
80
Chapter 2 Daulaut, W.E. & Clymo, R.S. (1998) Effects of temperature and watertable on the efflux of methane from peatland surface cores. Atmospheric Environment, 32, 32073218. Dise, N. (1992) Winter fluxes of methane from Minnesota peatlands. Biogeochemistry, 17, 71-83. Field, E.M. (1981). Peatland ecology in the British Isles: a bibliography. Institute of Terrestrial Ecology (National Environment Research Council), Cambridge. Fowler, D., Hargreaves, K.J., Choularton, T.W., Gallagher, M.W., Simpson, T., & Kaye, A. (1996) Measurements of regional CH4 emissions in the UK using boundary layer budget methods. Energy Conversion and Management, 769-775. Fowler, D., Hargreaves, K.J., MacDonald, J.A., & Gardiner, B. (1995a) Methane and CO2 exchange over peatland and the effects of afforestation. Forestry, 68, 327-334. Fowler, D., Hargreaves, K.J., Skiba, U., Milne, R., Zahniser, M., B, M.J., Beverland, I.J., & Gallagher, M.W. (1995b) Measurements of CH4 and N2O fluxes at the landscape scale using micrometeorological methods. Philosophical Transactions of the Royal Society of London, A, 339-356. Freeman, C., Nevison, G.B., Kang, H., Hughes, S., Reynolds, B., & Hudson, J.A. (2002) Contrasted effects of simulated drought on the production and oxidation of methane in a mid-Wales wetland. Soil Biology and Biochemistry, 34, 61-67. Gallagher, M.W., Choularton, T.W., Bower, K.N., Stromberg, I.M., Beswick, K.M., Fowler, D., & Hargreaves, K.J. (1996) Measurements of methane fluxes on the landscape scale from a wetland area in North Scotland. Atmospheric Environment, 28, 2421-2430. Garnett, M.H. (1998) Carbon storage in Pennine moorland and response to change. PhD Thesis, University of Newcastle-Upon-Tyne, Newcastle-Upon-Tyne. Garnett, M.H., Ineson, P., & Stevenson, A.C. (2000) Effects of burning and grazing on carbon sequestration in a Pennine blanket bog, UK. The Holocene, 10, 729-736.
81
Chapter 2 Gates, S. (2002) Review of methodology of quantitative reviews using meta-analysis in ecology. Journal of Animal Ecology, 71, 547-557. Gauci, V., Dise, N., & Fowler, D. (2002) Controls on suppression of methane flux from a peat bog subjected to simulated acid rain sulfate deposition. Global Biogeochemical Cycles, 16, 1-12. Gorham, E. (1991) Northern peatlands: role in the carbon cycle and probable responses to climatic warming. Eccological Applications, 1, 182-195. Grace, J. (2004) Understanding and managing the global carbon cycle. Journal of Ecology, 92, 189-202. Grace, J., Meir, P., & Malhi, Y. (2001). Keeping track of carbon flows between biosphere and atmosphere. In Ecology: achievement and challenge (eds M.C. Press, N.J. Huntly & S. Levin), pp. 249-269. Blackwell Science Ltd, Oxford. Guo, L.B. & Gifford, R.M. (2002) Soil carbon stocks and land use change: a meta analysis. Global Change Biology, 8, 345-360. Hargreaves, K.J. & Fowler, D. (1998) Quantifying the effects of water table and soil temperature on the emission of methane from peat wetland at the field scale. Atmospheric Environment, 32, 3275-3282. Hargreaves, K.J., Milne, R., & Cannell, M.G.R. (2003) Carbon balance of afforested peatland in Scotland. Forestry, 76, 299-317. Hughes, S., Dowrick, D.J., Freeman, C., Hudson, J.A., & Reynolds, B. (1999) Methane emissions from a gully mire in mid-Wales. U.K. under consecutive summer water table drawdown. Environmental Science and Technology, 33, 362-365. IPCC (1996) Climate Change 1995 – The science of climate change. Contribution of Working Group 1 to the Second Assessment Report of the Intergovernmental Panel on Climate Change Cambridge University Press, Cambridge.
82
Chapter 2 Joabsson, A., Christensen, T.R., & Wallen, B. (1999) Vascular plant controls on methane emissions from northern peat forming wetlands. Trends in Evolution and Ecology, 14, 385-388. Johnson, D.W. & Curtis, P.S. (2001) Effects of forest management on soil C and N storage: meta analysis. Forest Ecology and Management, 140, 227-238. Keppler, F., Hamilton, J.T.G., Braß, M., & Röckman, T. (2006) Methane emission from terrestrial plants under aerobic conditions. Nature, 439, 188-191. Komulainen, V.M., Tuittila, E.S., Vasander, H., & Laine, J. (1999) Restoration of drained peatlands in southern Finland: initial effects on vegetation change and CO2 balance. Journal of Applied Ecology, 36, 634-648. Lindsay, R.A. (1995) Bogs: the Ecology Classification and Conservation of Ombrotrophic Mires. Scottish Natural Heritage, Battleby. Lloyd, D., Thomas, K.L., Benstead, J., Davies, K.L., Lloyd, S.H., Arah, J.R.M., & Stephen, K.D. (1998) Methanogenesis and CO2 exchange in an ombrotrophic peat bog. Atmospheric Environment, 32, 3229-3238. Long, S.P., Ainsworth, E.A., Rogers, A., & Ort, D.R. (2004) Rising atmospheric carbon dioxide: Plants face the future. Annual Review of Plant Biology, 55, 591-628. MacDonald, J.A., Fowler, D., Hargreaves, K.J., Skiba, U., Leith, I.D., & Murray, M.B. (1998) Methane emission rates from a northern wetland; response to temperature water table and transport. Atmospheric Environment, 32, 3219-3227. Manley, J., Van Kooten, G.C., Moeltner, K., & Johnson, D.W. (2005) Creating carbon offsets in agriculture through no-till cultivation: A meta-analysis of costs and carbon benefits. Climatic Change, 68, 41-65. Martikainen, P.J., Nykeanen, H., Crill, P., & Silvola, J. (1992) The effect of changing water table on methane fluxes at two Finnish mire sites. Suo, 43, 237-240.
83
Chapter 2 Moncrieff, J.B., Beverland, I.J., O'Neill, D.H., & Cropley, F.D. (1998) Controls on trace gas exchange observed by a conditional sampling method. Atmospheric Environment, 32, 3265-3274. Nedwell, D.B. & Watson, A. (1995) CH4 production, oxidation and emission in a UK ombrotrophic peat bog: influence of SO-24 from acid rain. Soil Biology and Biochemistry, 27, 893-903. Oechel, W.C., Hastings, S.J., Vourlitis, G.L., Jenkins, M., Reichers, G., & Grulke, N. (1993) Recent changes of Arctic tundra from a net carbon dioxide sink to a source. Nature, 361, 520-523. Ogle, S.M., Breidt, F.J., & Paustian, K. (2005) Agricultural management impacts on soil organic carbon storage under moist and dry climatic conditions of temperate and tropical regions. Biogeochemistry, 72, 87-121. Oliver, H.R., Bell, B.G., & Clymo, R.S. (1998) The TIGER programme. Atmospheric Environment, 32, 3205-3205. Osenberg, C.W., Sarnelle, O., & Goldberg, D.E. (1999) Meta-analysis in ecology: Concepts, statistics, and applications. Ecology, 80, 1103-1104. Palmer, A.R. (2000) Quasireplication and the contract of error: Lessons from sex ratios, heritabilities and fluctuating asymmetry. Annual Review of Ecology and Systematics, 31, 441-480. Peterson, A.G., Ball, J.T., Luo, Y.Q., Field, C.B., Reich, P.B., Curtis, P.S., Griffin, K.L., Gunderson, C.A., Norby, R.J., Tissue, D.T., Forstreuter, M., Rey, A., & Vogel, C.S. (1999) The photosynthesis leaf nitrogen relationship at ambient and elevated atmospheric carbon dioxide: a meta-analysis. Global Change Biology, 5, 331-346. Roberts, K.A., Dixon-Woods, M., Fitzpatrick, R., Abrams, K.R., & Jones, D.R. (2002) Factors affecting uptake of childhood immunisation: a Bayesian synthesis of qualitative and quantitative evidence. Lancet, 360, 1596-1599.
84
Chapter 2 Shaw, S.C., Wheeler, B.D., Kirby, P., Phillipson, P., & Edinutids, R. (1996). Literature review of the historical effects of burning and grazing of blanket bog and upland wet heath, Rep. No. English Nature Research Reports No. I72. English Nature, Peterborough. Steiner, G. (1997). The bogs of Europe. In Conserving Peatlands (eds L. Parkyn, R.E. Stoneman & H.A.P. Ingram), pp. 25-34. CAB International, Wallingford. Tuittila, E.-S. (2000) Restoring vegetation and carbon dynamics in a cut-away peatland. PhD Thesis, University of Helsinki, Helsinki. van Kooten, G.C., Eagle, A.J., Manley, J., & Smolak, T. (2004) How costly are carbon offsets? A meta-analysis of carbon forest sinks. Environmental Science & Policy, 7, 239-251. Waddington, J.M., Roulet, N.T., & Swanson, R.V. (1996) Water table control of CH4 emission enhancement by vascular plants in boreal peatlands. Journal of Geophysical Research, 101, 22775-22785. Wang, X.Z. & Curtis, P. (2002) A meta-analytical test of elevated CO2 effects on plant respiration. Plant Ecology, 161, 251-261. Whiting, G.J. & Chanton, J.P. (1993) Primary production control of methane emission from wetlands. Nature, 364, 794-795.
85
Chapter 3 Chapter 3: Blanket Bog Site Characteristics and the Role of Management 3.1 Introduction Determinants of blanket bog vegetation are climatic (Moore & Bellamy, 1974; Lindsay et al., 1988; Lindsay, 1995) and anthropogenic. There is some evidence that human destruction of forest since the last glaciation led to the formation of blanket bog in some areas (Moore & Bellamy, 1974; Jacobi et al., 1976; Moore et al., 1984; Tallis, 1991; Moore, 1993) and there are demonstrable links between modern anthropogenic management and blanket bog vegetation (Chapter 1). There are likely to be interactions between management actions and climate. Therefore, if we are to understand ecosystem response to climate change or the effect of ecosystems on the climate system, then the implications of management actions on that ecosystem need to be understood. Further, if we are to mitigate for any negative consequences on the climate through management practice resulting in a positive global warming potential, then it is only through changes to management that this can be redressed. The UKCIP02 report predicts warmer winters and drier summers (Hulme et al., 2002) if these predictions are realised then these will impact on the vegetation of blanket bog. Management practices such as burning and grazing have been practiced on UK peatlands for centuries (Chapter 1). Therefore an understanding of the impacts of management is vital for predictions of climatic change vegetation response. 3.2 Study aims Here data from northern England and the north of Scotland are used to explore how management affects the vegetation of blanket bog. Management is investigated through vegetation survey of a replicated split plot management experiment and sites with gradations in regular management practices. Attempts are made to separate out innate site characteristics from those identifiable to management. The implications of management on carbon fluxes are explored in Chapter 5.
86
Chapter 3 3.3 Methods 3.3.1 Site Descriptions 3.3.1a Moor House
Figure 3.1: Locations of Hard Hill experimental plots and location of Moor House in the UK with peat over 50 cm marked as black (inset adapted from Lindsay 1995). Map reproduced by kind permission of Ordnance Survey © Crown Copyright. Moor House is situated in the Northern Pennines (Grid Ref NY 757 328), has an area of 74 km2 and ranges in altitude from 290 to 850 m asl. It is a large part of the catchment of the River Tees and a National Nature Reserve (NNR), a UNESCO Biosphere Reserve and a European Special Protection Area. The site includes exposed summits, extensive blanket peatlands, upland grasslands, pastures, hay meadows and deciduous woodland. Moor House has history of scientific research stretching back to the early 1950’s and has a number of long-term experiments including investigation of management on blanket bog at Hard Hill. This is a split block, burning and grazing experiment established 1954. The Hard Hill site is located on blanket peat of approximately 1-2 m depth, mean annual rainfall is approximately 1900 mm with mean temperature of 5.1 oC (Heal & Smith, 1978).
87
Chapter 3 Figure 3.1 and Table 3.1 detail location, vegetation and management of the Moor House and the Hard Hill site. The entire study area was burned prior to the construction of the experimental blocks in 1954 and the method used, was and still is, similar to traditional moorland burning (Hobbs & Gimingham, 1987). The site is arranged as in Figure 3.2 with four blocks two grazing treatments, grazed and ungrazed and three burning regimes; 0 burn (burnt in 1954 only) 10 year and 20 year rotational burning. Grazing is light with approximately 0.02-0.2 ewes per hectare (Smith & Forrest, 1978).
Figure 3.2: Details of Hard Hill experimental set up (Adamson & Kahl, 2003). 3.3.1b Forsinard The Forsinard and Dorrery RSPB Nature Reserve (Grid Ref NC 905 465) is located in Sutherland, Scotland and covers an area of 112 km2 and ranges from 44 to 580 m above sea level (asl), with most of the deep peatlands between 120 and 438 m asl. Field-work was conducted at a total of nine ombrotrophic blanket bog sites between Grid Ref NC 83 45 in the west and NC 97 45 in the east (Figure 3.3). Location and
88
Chapter 3 management details of each of these sites are given in Table 3.1. The climate of the area is characterised by high and frequent rainfall with annual amounts in the region of 1000-1500 mm yr-1 with approximately 160 - 180 wet days yr-1 (a 24 hour period where over 1 mm rainfall is recorded) (Lindsay et al., 1988). Mean daily temperatures are in the region of 8 oC. The reserve lies in a bioclimatic region considered to be Euoceanic, very humid, southern boreal and lower oroboreal and the major area of peat formation in the flow country conforms to this classification (Birse, 1971, cited in Lindsay et al., 1988). The reserve forms part of the Peatlands of Caithness and Sutherland, an internationally important peatland habitat recognised by status as a Ramsar site, Special Protection Area (SPA), candidate Special Area of Conservation (SAC) and proposed World Heritage Site.
Figure 3.3: Locations of sampling sites in relation to Forsinard Sutherland and location of Forsinard in the UK with peat over 50 cm marked as black (inset adapted from Lindsay 1995). Map reproduced by kind permission of Ordnance Survey © Crown Copyright.
89
Chapter 3 Table 3.1: Details of Hard Hill Site at Moor House NNR and the 11 sampling sites located within the Forsinard Reserve. Determination of National Vegetation Communities (NVC) (Rodwell, 1991) was aided by the use of ComKey computer software (Legg, unpublished). All NVC communities constitute Biodiversity Action Plan (BAP) priority habitats under present JNCC guidelines. Low Med High Deer inferred from an RSPB survey of animal footprints see Appendix. *NVC of vegetation derived from Calluneto - Eriophoretum (Eddy et al., 1969) Site
Grid Ref
Hard Hill: Block A Block B Block C Block D Nam Breac
Alt NVC Community/sub-community General management and (m a.s.l.) site characteristics *M19b Calluna vulgaris Eriophorum Nature conservation, NY 743330 600-630 vaginatum blanket raised experimental plots with NY 740330 mire/Empetrum nigrum sub grazing and burning NY 736330 community treatments. NY 738331 NC 831 451 190 M17b Scirpus cespitosus Eriophorum Nature conservation, high vaginatum mire/Cladonia sub deer. Bare peat evident community throughout site
Sletill
NC 933 456 185
M17a Scirpus cespitosus Eriophorum Nature conservation, low vaginatum mire/Drosera-Sphagnum deer. Relatively intact site sub community
Leir
NC 958 461 195
Maol Donn
NC 975 454 165
Fire Site
NC 881 501 105
Site L
NC 861 467 180
M17a Scirpus cespitosus Eriophorum Nature conservation, low vaginatum mire/Drosera-Sphagnum deer. Relatively intact site sub community though some bare peat present M18a Erica tetralix Sphagnum Nature conservation, low papillosum raised and blanket deer. Relatively intact site mire/Sphagnum magellanicum Andromeda polifolia sub community M15 Scirpus cespitosus Erica tetralix Not within reserve wet heath boundary, open for sheep and deer stalking. Fire burnt early 2004, burnt and unburnt areas within the same site M17a Scirpus cespitosus Eriophorum Nature conservation, low vaginatum mire/Drosera rotundifolia deer. Bare peat evident Sphagnum spp. sub community throughout site
Site M
NC 856 444 220
M17a Scirpus cespitosus Eriophorum Nature conservation, high vaginatum mire/Drosera rotundifolia deer. Bare peat evident Sphagnum spp. sub community throughout site
Site N
NC 843 447 180
M17a Scirpus cespitosus Eriophorum Nature conservation, high vaginatum mire/Drosera rotundifolia deer. Bare peat evident Sphagnum spp. sub community throughout site
Cross Lochs NC 864 465 180 Drains
M17a Scirpus cespitosus Eriophorum Nature conservation, med vaginatum mire/Drosera-Sphagnum deer. Drained site, blocked sub community and unblocked drains sampled. Drains cut in the 1970's and 80's and blocked 1/08/96
90
Chapter 3 Table 3.2: Number of relevés per site and dates of vegetation sampling from Forsinard sites 2004-2005. Main Site
Site
No. vegetation relevés
Vegetation sampling dates
Moor House
Hard Hill
3 per plot
May 2002
18 per block Forsinard Reserve
Nam Breac
20
July – Aug 2004
Sletill
20
July – Aug 2004
Leir
20
July – Aug 2004
Maol Donn
20
July – Aug 2004
Fire
15 burnt
Aug 2004
15 unburnt Site L
20
July – Aug 2004
Site M
20
July – Aug 2004
Site N
20
July – Aug 2004
Cross Lochs Drains
15
July 2005
3.3.2 Vegetation Characterisation Field-work began at Moor House in May 2002 and at Forsinard in July 2004, details of sampling dates are given in Table 3.2. 3.3.2a Moor House In each of the split plots three random, 0.32 m2 relevés (same area as gas flux chambers, see Chapter 4) were sampled. The visual percentage cover of all species including vascular plants, bryophytes, macro-lichens and bare peat was recorded. 3.3.2b Forsinard Vegetation sampling began in July 2004 and was initially completed in August 2004 except for the Cross Lochs Drain site, which was sampled in June of 2005. At each site the vegetation composition and structure was recorded in the following way:
The visual percentage cover of all species including vascular plants, bryophytes, macro-lichens and bare peat was recorded from relevés as above.
Deer, sheep and hare, faecal count by species within relevés.
Deer and sheep footprint count by species within relevés.
91
Chapter 3
Vegetation canopy height and structure, using the percent obscured stick method, which is as follows: A stick marked with bands of 2 cm was placed at nine points in the relevé in a 3 x 3 grid. The height of the moss layer and any other species touching the stick and within 5 cm of the stick, are recorded with the stick held vertically at arms length. The visual percentage of the stick that is obscured by the vegetation in each 2 cm band is then recorded.
Site surface (< 10 cm) pH was measured with 15 replicates per site.
The Bush recording soil penetrometer (Campbell & O'Sullivan, 1991) was used for pressure readings at every 1 cm to a depth of 50 cm, with 50 insertions per site except at the Cross Lochs Drains. Penetrometer readings at the Cross Lochs Drains were taken from five 10 m transects from unblocked and blocked drains insertions were at 0.5m and every metre from 1 –10 m.
3.3.3 Statistical Analysis Vegetation data were analysed using Detrended Correspondence Analysis (DCA) and vegetation and environmental variables with Canonical Correspondence Analysis (CCA) and Redundancy Analysis (RDA). DCA, CCA and RDA were implemented in Canoco 4.5 software. The percent obscured stick method data were analysed to give indices of shrub biomass, canopy height, density and heterogeneity (G. M. Davies unpublished) using PObscured computer software (Legg, unpublished). PObscured calculates the logit regression of the percentage obscured in each band against height, means and standard deviations are then computed for each quadrat from the nine stick observations. The calculated indices are as follows (Legg, unpublished):
10% height and 50% height. The height on the stick at which 10% and 50% of the particular band is obscured. These data are obtained by fitting the logistic curve to the data and interpolating (or extrapolating) from the smoothed curve. These data should be more robust measures of canopy height (though 50% can be negative for very thin crowns) than simple height measurements as these are more variable and prone to extreme values.
Volume. Volume is the area between the fitted curve and zero height. It is called 'volume' because it is derived from a height times an area (%
92
Chapter 3 obscured). It is used as an index of biomass and may be expected to give good correlation although this has yet to be confirmed in vegetation other than Callunetum (G. M. Davies unpublished).
Intercept and Slope: These are the intercept and slope of the logistic regression of percent obscured on height. The intercept is the logit of percent obscured extrapolated to the base of the stick reflecting light penetration to ground level, and the slope is the increase in logit (percent obscured) per cm increase in height reflecting canopy density.
3.3.4 Community comparison Comparison of vegetation data with the NVC and the communities of the Moor House reserve (Eddy et al., 1969) was done using ComKey (Legg, Unpublished). The communities of Eddy et al., (1969) included in this analysis are the CallunetoEriophoretum Typical facies, Calluneto-Eriophoretum Sphagnum recurvum facies, Calluneto-Eriophoretum Empetrum nigrum facies, Calluneto-Eriophoretum Burnt facies, Trichophoretum-Eriophoretum typical facies and Eriophoretum high level facies. The Calluneto-Eriophoretum community is considered synonymous with M19 Calluna vulgaris-Eriophorum vaginatum blanket mire, the TrichophoretumEriophoretum typical facies with M18 Erica tetralix Sphagnum papillosum raised and blanket mire and Eriophoretum high level facies synonymous with the M20 Eriophorum vaginatum blanket and raised mire, NVC communities (Rodwell, 1991). Eddy et al., (1969) originally mapped the Hard Hill site as the CallunetoEriophoretum burnt facies. Two approaches are used, firstly simple classification of treatments to a community by reference to Rodwell (1991) and using the Czekanowski similarity coefficient, commonly used by vegetation consultants using e.g. MAVIS (Smart, 2000). Secondly by deriving a Presence-Weighted Similarity (PWS) and Sørensen Similarity coefficients for relevés to data from Eddy et al., (1969) and tabulated NVC samples and then analysed using Principal Components Analysis (PCA). Communities selected for use in PCA were those that matched with a similarity of greater than 50 using PWS. The Czekanowski, PWS and Sørensen coefficients are defined as (Legg, unpublished):
93
Chapter 3 Czekanowski = 2 bm / (S+C) where: bm = minimum of the abundance in the sample and community S = number of species in sample C = number of species in community PWS = (sum (bp) / 5*S) * 100 where: bp = community presence values of species occurring in both relevé and community, and S = number of species in relevé.
Sørensen = 2 B / (S+C) where: S = number of species in sample C = number of species in community B = number of species that occur in both sample and community Czekanowski is a symmetrical coefficient that assumes that the sample and the type community are equivalent in every way. Thus the match will tend to be biased towards species-poor type communities that have a similar total number of species to the sample. Similarly, it is not appropriate to compare cover-abundance scores of the sample with presence classes of the type and is not therefore suitable for single relevé data. PWS is the sum of NVC community table frequency values (1-5) for only species that occur in both the relevé and the community, divided by 5 multiplied by the number of species in the relevé, multiplied by 100. This will give 100 for community containing all S species with presence class 5, or 20 for all species present with presence class 1. The score is thus heavily weighted towards the most frequent species.
94
Chapter 3 Sørensen is a symmetrical coefficient that assumes that the sample and the type community are equivalent in every way. Thus the match will tend to be biased towards species-poor type communities that have a similar total number of species to the sample. All other summary statistics and graphical plots were generated using Minitab 13 and Microsoft Excel 2000 software.
95
Chapter 3 3.4 Results 3.4.1 Moor House Table 3.3 shows the species recorded from a total of seventy two relevés in the vegetation survey of Hard Hill, number of relevés for each species in each treatment and species codes for ordination diagrams. There are a total of twenty-five species that include nine vascular plants, nine mosses, three liverworts and four lichens. All species are common to mire and heathland habitats, none restricted or rare in the UK. Table 3.3: Species, Species code, and total number of relevés in each treatment for each species recorded from a total of 72 relevés sampled from Hard Hill experimental site, Moor House NNR. Species are arranged in order of abundance in terms of the total number of relevés they are present in. Species Species code Grazed Ungrazed Calluna vulgaris (L.) Hull Call vul 36 35 Eriophorum vaginatum L. Erio vag 34 35 Eriophorum angustifolium Honck. Erio ang 30 23 Dicranum scoparium Hedw. Dic scop 24 19 Rubus chamaemorus L. Rub cha 21 18 Hypnum jutlandicum Holmen & E.Warncke Hyp jut 14 12 Sphagnum capillifolium (Ehrh.) Hedw. Sph cap 10 13 Calypogeia muelleriana (Schiffn.) Müll.Frib. Caly mue 15 7 Empetrum nigrum subsp. nigrum L. Emp nig 7 13 Polytrichum commune Hedw. Poly com 8 11 Plagiothecium undulatum (Hedw.) Bruch, Plag und 8 11 Schimp. & W.Gümbel Lophocolea bidentata (L.) Dumort. Loph bid 7 11 Cladonia portentosa (Dufour) Coem. Clad imp 5 6 Pleurozium schreberii (Brid.) Mitt. Pleu sch 5 4 Aulacomnium palustre (Hedw.) Schwägr. Aula pal 6 Mylia taylorii (Hook.) Gray Myl tay 3 3 Vaccinium vitis-idaea L. Vacc vit 2 3 Cladonia chlorophea (Flörke ex Sommerf.) Clad chl 1 4 Sprengel. Vaccinium myrtillus L. Vacc myr 2 3 Cladonia sp. Clad sp 1 2 Dryopteris dilatata (Hoffm.) A. Gray Dry dil 1 1 Rhytididelphus loreus (Hedw.) Warnst. Rhy lor 2 Hypogymnia physoides (L.) Nyl. Hyp phy 1 1 Sphagnum fallax (H.Klinggr.) H.Klinggr. Sph rec 1 0 Trichophorum cespitosum (L.) Hartm. Trich ces 1 0
96
0 burn 24 24 11 9 15 21 5 9 -
10 yr 20 yr 23 24 24 21 18 24 16 18 14 10 2 3 13 5 12 10 7 4 12 7
11 3 8 8 3 0 2
2 6 2 2 3 3
6 9 1 1 1 3 -
2 1 1 2 1 -
2 4 1 1 1 -
1 1 2 1
Chapter 3 Figure 3.4 and 3.5 show sample and species plots of a DCA using the Hard Hill vegetation data. The longest gradient length is 2.8 and axes 1 and 2 account for 17.2 % and 10.9 % of the variation in the vegetation data, where as the 3rd and 4th axes account for 8 and 6.7 % respectively. Sample 36 is an outlier due to the abundance of T cespitosum removing this sample gives gradient length of the first axis is for 3.1 and axes 1 and 2 account for 16 % and 10.6% of the variation in the vegetation data and the 3rd and 4th axes account for 6.9 and 5.1 % respectively. The gradient lengths are relatively short and therefore linear techniques such as RDA are appropriate (Lepš & Šmilauer, 2003).
3.0
29
23 2256 30 42 9
24
28
40
57
67 61 55 68 41 69 37 8 11 18 38 52 60 66 39 53 17 54 10 34 35 65 12 3 59 2 46 15 5 47 745433172 50 48 64 253216 14 51 63 1 586 27 70 49 62
36
26
33 21 71
13
44 4 20
-0.5
19
-0.5
3.0
Figure 3.4: Axes 1 and 2 of a DCA of species percentage cover data showing samples from Hard Hill. Plot codes are as in Table 3.3. Axes 1 and 2 accounted for 17.2 % and 10.9 % respectively of total variation in vegetation data.
97
4
Chapter 3
Hyp jutl Aula pa Hypo phy Plag und Clad imp Call vul Clad chl Rub cham
Trich ce Dicr sco
Pleu sch Loph bid Erio vag
Erio ang Caly mue
Sph recu Rhy loreVacc Dry dila myr Emp Cladnigr squ Myli tay Sph cap
-1
Vacccom vit Poly
-1
4
Figure 3.5: Axes 1 and 2 of a DCA of species percentage cover data showing species from Hard Hill. Plot codes are as in Table 3.3. Axes 1 and 2 accounted for 17.2 % and 10.9 % respectively of total variation in vegetation data. Figure 3.6 shows a species ordination diagram of an RDA of the seventy two relevés from the Hard Hill data, only the fifteen most abundant species are depicted. Axes 1 and 2 accounted for 21.4 % and 5.6 % respectively of total variation in vegetation data and 77.2 % and 20% respectively of species-environment relationship. Restricted Monte Carlo permutation test according to the split plot structure of the experiment revealed the first axis to be highly significant (p
